Phẫu Thuật Gân Bàn Tay (Tendon Surgery of the Hand)

238
Section 2:  Primary Flexor Tendon Surgery
increase the contact area of fixation and to allow bio­
logical healing of the periosteum to the phalanx. The
extensor retinaculum graft is fixed to the floor of the
sheath with both sutures passing through the perios­
teum and sutures to the rims of the sheath, so that the
load can be distributed over a greater area of fixation.
Although this method is more demanding technically
and the donor site of the extensor retinaculum is aes­
thetically less ideal, we obtained good functional out­
comes of the fingers and high patient satisfaction.
less frictional resistance to tendon gliding.23 We know
nothing about the fate of free tendon grafts in recon­
structed pulleys. The grafted tendons likely remain avas­
cular and probably are embedded in scar.
At surgery, it is very important to fix the graft ade­
quately to withstand high loads. Interweaving the
tendon graft to the remnants of the sheath is difficult if
the site of pulley rupture is at the side and close to the
attachments of the sheath. We harvested a periosteal
strip connecting with the extensor retinaculum to
References
1. Doyle JR: Anatomy of the finger flexor tendon sheath and
pulley system, J Hand Surg (Am) 13:473–484, 1988.
2. Peterson WW, Manske PR, Bollinger BA, et al: Effect of pulley
excision on flexor tendon biomechanics, J Orthop Res 4:96–
101, 1986.
3. Bowers WH, Kuzma GR, Bynum DK: Closed traumatic
rupture of finger flexor pulleys, J Hand Surg (Am) 19:782–
787, 1994.
4. Arora R, Fritz D, Zimmermann R, et al: Reconstruction of the
digital flexor pulley system: A retrospective comparison of
two methods of treatment, J Hand Surg (Eur) 32:60–66, 2007.
5. Bollen SR: Injury to the A2 pulley in rock climbers, J Hand
Surg (Br) 15:268–270, 1990.
6. Gabl M, Reinhart C, Lutz M, et al: The use of a graft from the
second extensor compartment to reconstruct the A2 flexor
pulley in the long finger, J Hand Surg (Br) 25:98–101, 2000.
7. Schweizer A: Biomechanical properties of the crimp grip posi­
tion in rock climbers, J Biomech 34:217–223, 2001.
8. Vigouroux L, Quaine F, Labarre-Vila A, et al: Estimation of
finger muscle tendon tensions and pulley forces during spe­
cific sport-climbing grip techniques, J Biomech 39:2583–2592,
2006.
9. Marco RA, Sharkey NA, Smith TS, et al: Pathomechanics of
closed rupture of the flexor tendon pulleys in rock climbers,
J Bone Joint Surg (Am) 80:1012–1019, 1998.
10. Vigouroux L, Quaine F, Paclet F, et al: Middle and ring fingers
are more exposed to pulley rupture than index and little
during sport-climbing: A biomechanical explanation, Clin
Biomech (Bristol, Avon) 23:562–570, 2008.
11. Lin GT, Amadio PC, An KN, et al: Biomechanical analysis of
finger flexor pulley reconstruction, J Hand Surg (Br) 14:278–
282, 1989.
12. Bunnell S, Böhler J: Die Chirurgie der Hand, Wein, Wilhlem
Maudrich Verlag, Teil 1:533–643, 1958.
13. Kleinert HE, Bennett JB: Digital pulley reconstruction employ­
ing the always present rim of the previous pulley, J Hand Surg
(Am) 3:297–298, 1978.
14. Lister GD: Reconstruction of pulleys employing extensor reti­
naculum, J Hand Surg (Am) 4:461–464, 1979.
15. Bader KF, Sethi G, Curtin JW: Silicone pulleys and underlays
in tendon surgery, Plast Reconstr Surg 41:157–164, 1968.
16. Widstrom CJ, Johnson G, Doyle JR, et al: A mechanical
study of six digital pulley reconstruction techniques: Part I.
Mechanical effectiveness, J Hand Surg (Am) 14:821–825,
1989.
17. Klauser A, Bodner G, Frauscher F, et al: Finger injuries in
extreme rock climbers. Assessment of high-resolution ultra­
sonography, Am J Sports Med 27:733–737, 1999.
18. Bodner G, Rudisch A, Gabl M, et al: Diagnosis of digital flexor
tendon annular pulley disruption: Comparison of high fre­
quency ultrasound and MRI, Ultraschall Med 20:131–136,
1999.
19. Buck-Gramcko D, Dietrich FE, Gogge S: Evaluation criteria in
follow-up studies of flexor tendon therapy, Handchirurgie
8:65–69, 1976.
20. Moutet F: Flexor tendon pulley system: Anatomy, pathology,
treatment, Chir Main 22:1–12, 2003.
21. Karev A: The “belt loop” technique for the reconstruction of
pulleys in the first stage of flexor tendon grafting, J Hand Surg
(Am) 9:923–924, 1984.
22. Okutsu I, Ninomiya S, Hiraki S, et al: Three-loop technique
for A2 pulley reconstruction, J Hand Surg (Am) 12:790–794,
1987.
23. Nishida J, Amadio PC, Bettinger PC, et al: Flexor tendon-
pulley interaction after pulley reconstruction: A biomechani­
cal study in a human model in vitro, J Hand Surg (Am)
23:665–672, 1998.

239
CHAPTER
21 
THE EVOLUTION OF
END-TO-END SURGICAL
TENDON REPAIRS
Robert Savage, MB, FRCS, FRCS Ed Orth, MS
of various important tissues, for example, intestine,
bladder, abdominal wall, artery and vein graft, etc., I was
taught that the more suture material that crossed the
repair site, the greater was the repair strength and distri­
bution of load (and prevention of leakage) and the
more assured would be the subsequent result.
I was taught tendon repair by using a two-strand
Bunnell or Kessler type repair and was told the outcome
of repair was likely to be complicated by a high inci­
dence of repair dehiscence, tendon adhesions, and stiff­
ness. I remember thinking that the repair did not match
the general closure principles I had learned.
Tendon repair has changed significantly since then
and I would say that now it is accepted in many hand
centers that a multistrand repair is required to allow
protected movement during the early stages of healing,
so critical for preventing adhesions, and that in most
circumstances this will result in satisfactory movement,
a low incidence of rupture, and a good functional
outcome.
EARLY REPAIRS AND SURGICAL STRATEGY
In 1922, Bunnell1 described the principles of tendon
repair that we would find it hard to disagree with now.
His details and materials then were slightly different to
ours now, but the idea then was similar. He described a
robust criss-cross suture, in the anterior half of the
tendon to protect the tendon blood supply, sufficiently
strong to allow early movement, with little exterior
exposure of suture material to reduce adhesion, a splint
with the wrist flexed but the fingers free so that “the
muscles that pull the tendons are thus robbed of their
power of too much traction and still they can keep the
tendon actively moving.” Cooperation by the patient
was essential.
Just a year later in 1923, Lahey2 reported a method
of tendon repair that used a grasping technique where
the suture material encircled some of the tendon fibers,
gripping them tightly to prevent slippage. Three addi­
tional interrupted sutures at the tendon junction were
added to promote accurate apposition. The sutures were
overtightened sufficient to allow the suture to settle in
OUTLINE
Approximately 100 years ago, tendon repair for acute
division was performed with silk sutures using two-
strand repair methods, and although it is a gross
simplification, mediocre results with resultant poor
movement and function were attributed to tissue
adhesion and poor capacity for tendons to heal. The
problem was countered by strategies such as tendon
grafting or excision of the tendon sheath, but ulti­
mately better methods of direct repair have evolved. It
has been realized that mobilization of the finger and
the injured tendon prevents adhesion of the tendon
to tendon sheath and to bone and that the necessary
post repair mobilization can most safely be carried out
using multistrand tendon repair techniques. Current
practice divides the tendon repair into core sutures
and peripheral sutures, which together contribute
greatly to repair strength and gap resistance. Tough
synthetic suture materials that have low tissue reactiv­
ity and that retain their initial strength are now
standard. Such a strategy of robust repair and active
protected postsurgery mobilization ensures a high
chance of obtaining clinically excellent movement and
function. In addition, the worst results that follow
partial repair failure with gap formation, and total
failure with repair rupture, are infrequent.
There have been many influences on tendon repair over
the past century related to the history of repair methods,
to our understanding of the biology and healing of
tendons, to laboratory studies on repair techniques, to
availability of suture materials, and to the practice of
post surgical mobilization. There is a mountain of sci­
entific and clinical literature on this fascinating subject,
but I have drawn on a small part of it that appears most
relevant.
My impression when I commenced tendon surgery
as an orthopedic trainee in 1983 was to relate what I
saw of tendon repair as instructed, compared to other
aspects of surgery that I had learned, mainly as a general
surgery trainee in the United Kingdom. In the closure

240
Section 2:  Primary Flexor Tendon Surgery
suture materials of fine caliber, 5-0 for the core repair
and 6-0 or 7-0 for the peripheral running suture. The
sheath was opened sufficiently to complete the repair
but was retained enough to prevent bowstringing.
Results from primary repair were improved but signifi­
cant numbers of tenolysis were required. It would be
interesting to speculate whether there would have been
advantage if they had used the probably superior Mason-
Allen or original Kessler core repair.
By 1981, Kleinert et al8 had changed to a modified
Kessler core suture of 3-0 or 4-0 braided polyester
together with a 6-0 nylon epitendon suture. Both these
repair elements persisted for most of the next 25 years
mainly for the good, and there was widespread develop­
ment of hand therapy services that were an essential part
of Kleinert’s method.
PATHOLOGY OF TENDON HEALING
Our understanding of tendon healing capability has
strongly influenced clinical practice over the decades,
sometimes for the better and sometimes for the worse.
In 1932, Mason and Shearon,9 studying dog tendons,
found that sheath tissues proliferated, that tendon ends
frequently separated, and that the resultant gap became
filled with blood and healing tissue that grew out from
the tendon ends. This suggested that the sheath should
be repaired and that the tendon should be moved.
Although this was practiced by Bunnell in 1922 and
Garlock in 1926,10 this practice did not seem to persist.
Other studies demonstrated that mobilized animal
tendons regained strength more quickly than immobi­
lized tendons.
The views of Peacock (1965)11 and Potenza (1969)12
dominated for a couple of decades. The widely held
view on tendon healing was that finger and sheath
tendons had very little capacity for healing. Peacock
popularized the “one wound” concept in which healing
to all tissues from skin down to bone was in the general
process of inflammation, granulation, and scar forma­
tion, such that expecting the tendon to move within the
scar appeared impractical.
Potenza’s view was possibly more negative than this,
for he concluded that tendon had no repair potential
itself and that healing only occurred when granulation
tissue grew from the neighboring tissues and tendon
sheath into the tendon, although following subsequent
work it seems their conclusions were misinterpreted. He
found that surgical pricking of the tendon induced
adhesions and that worse damage to the tendons pro­
duced more adhesion. Excision of the tendon sheath
caused no delay of healing and this practice was advo­
cated by Verdan.6
Matthews and Richards (1976)13 and Lundborg
(1976)14 showed that tendon proliferation was seen
in parts of rabbit tendon devoid of blood supply,
apparently as a result of nutrition by synovial fluid.
when the tendon was moved and active postoperative
finger mobilization was commenced. The suture mate­
rial was either silk or linen. Regrettably this work did
not contain clinical results but it would be fascinating
to try out this method today, for it seems not dissimilar
to some current methods and it used active mobiliza­
tion prior to a very long period where immobilization
or passive mobilization was used.
In 1944, Bunnell3 advocated overtightening the
sutures slightly to counteract the tendency for tendon
ends to separate and now he also immobilized the
finger for 3 weeks after surgery. He also popularized the
use of stainless steel for its low biological reaction and
high tensile strength. To counteract the tendency of wire
to break when repeatedly bent, he developed the pull-
out wire system. Bunnell had named zone 2 of the
fingers “no man’s land” for primary tendon repair. It is
not clear exactly what caused poor results in Bunnell’s
cases, although from the discussion that follows we
might assume it was a subtle mix of repair quality and
postsurgical immobilization. Tissue adhesion at the
repair site was recognized, and he popularized primary
tendon grafting, thus taking the site of surgical repair
away from zone 2 and placing it at the end of the finger
and within the palm or forearm.
In 1941, Mason and Allen4 devised a technique that
was quite similar to the subsequent original Kessler
repair in 1973.5 After the creation of an anchor point
on each side of the tendon the suture was passed trans­
versely across the tendon but proximal to the anchor
point and then tied to the corresponding suture from
the opposite tendon end. Additional sutures were placed
at the tendon margin. Animal studies showed that the
external sutures were soon covered by a thin sheet of
tissue, countering the argument that sutures on the
tendon surface should be avoided because they caused
adhesions. Again, reactive silk was used and there were
multiple knots, giving potential for weakness.
In 1960, Verdan6 described primary repair in zone 2
using a completely different strategy to counteract fre­
quently observed poor results from tendon grafting and
from the adhesions that followed attempts at primary
repair. His repair comprised two pins that transfixed the
tendon at a distance from the cut end and a fine arterial
type epitendinous suture. The sheath at the repair site
was excised for 1 inch so that the adhesions necessary
for healing were not attached to firm tissue. Results were
not particularly good by current standards but one of
his principles has stood the test of time, for the marginal
peripheral suture is widely used now.
In 1973, Kleinert et al7 described a simplified short
criss-cross core suture, a variant of the Bunnell longer
criss-cross suture, and a peripheral marginal running
suture. The principle was to produce a neat repair with
no gaps and then to apply the now well-known elastic
band dynamic mobilization system. They used synthetic

241
Chapter 21:  The Evolution of End-to-End Surgical Tendon Repairs
three such lock points (effectively producing a blanket
stitch down each side of the tendon). The study used
stainless steel sutures and sequential radiography to
produce photographs of the suture unraveling: under
very light tension (3 N) the curly lock points began to
unravel, at 9 N the lock points were nearly untwisted,
and at 15 N the lock points had straightened completely
and the suture migrated progressively to the cut tendon
end. The study showed no strength advantage to the
extra lock points and a stretch disadvantage to the mul­
tiple lock points, because there was more suture to
unravel and to create a wider gap. The authors con­
cluded that the “locking loop” should not be included
in designs for load-bearing tendon repair.
However, Kleinert popularized the very important
epitendon or peripheral suture, previously described by
Verdan, with the principal aim of creating a smooth,
neat repair. Wade et al (1989)18 investigated the mechan­
ical properties of the peripheral suture. Using a simpli­
fied Kessler core repair without locking loops, they
added a running “over and over” peripheral stitch
(Kleinert) using 5-0 braided polyester; and they studied
a Halsted-type peripheral repair using 5-0 polypropyl­
ene. Their values for initial gap forming under a load
test were core stitch alone, 3.4 N; core stitch with
running peripheral suture, 22 N; core stitch with Halsted
peripheral suture, 39 N.
Kitsis et al19 subsequently used the Halsted periph­
eral repair clinically and reported a large series with very
good results and a very low rupture rate. The Halsted
repair is easy to perform and the suture lies mainly
beneath the tendon surface.
Silfverskiöld and Andersson20 taught us that a periph­
eral cross-stitch repair (Figure 21-1) gave similar loads
and gap resistance to Wade’s Halsted-type repair (namely
about 50 N at a 2-mm gap and 60 N ultimate failure).
A minimum of 14 strands is necessary for best effect
according to the report of Kubota et al in 1996,21 but
In addition, Matthews and Richards (1974, 1976)15
designed a very clever biological model for tendon
healing experimentation. This followed a clinical obser­
vation by Harold Richards (personal communication)
that sometimes when a clinical case of tendon division
presented late, on exploring the tendon sheath, little by
way of adhesion and scarring was found around the
tendon end. They postulated that it was the factors that
surgeons apply to the finger that caused poor healing
and adhesion formation. The laboratory experiments
were carried out in rabbit flexor tendons. The ingenious
part of the methodical analysis was to partly divide the
tendon so that it could be observed within the tendon
sheath in an unsutured, non–sheath-injured, and non­
immobilized paw. This setup served as the control for
subsequent addition of surgical factors, namely sutur­
ing, sheath injury, and immobilization.
The control tendons showed evidence of healing
within the partial tendon gap and no sign of adhesions.
The three surgeon-induced factors each produced
modest adhesions between the tendon and the tendon
sheath, but these resolved with time. When two factors
were applied together, there were moderate adhesions
that eventually resolved. But when all three factors
were applied, the adhesions were dense, restrictive, and
unresolving.
These very important studies certainly changed the
view on tendon healing, although the authors did not
suggest how the surgical dilemma could be resolved. As
I see it, that came gradually with surgeons carrying on
doing what they thought should work clinically. I argue
in retrospect that these and many other laboratory
studies support the notion that if surgical repair can be
done neatly but with sufficient strength for immobiliza­
tion to be avoided, then healing can occur without
dense adhesions forming.
FURTHER DEVELOPMENT OF SURGICAL
REPAIR TECHNIQUE
In the 1970s and 1980s, most surgeons were using a
two-strand tendon repair technique. Various ways of
making the tendon suture weave into or grip onto the
tendon fibers were devised and published. It appears
that the principle laid down by Bunnell and Kessler, that
the suture gripped firmly onto tendon fibers, was partly
lost in Kleinert’s method of the foreshortened Bunnell
core suture and the modified Kessler-type core suture,
which appears to have less gripping power onto the
tendon than the original methods.
In 1991, Mashadi and Amis16 studied the modified
Kessler technique described by Pennington in 1979,17 in
which the transverse part of the suture passed superficial
to the longitudinal part, apparently creating a locked
arrangement of suture against tendon. This was com­
pared with Verdan’s technique, which had two of the
lock arrangements, and the Ketchum repair, which had
Figure 21-1  A 14-strand Silfverskiöld peripheral suture.
(Modified from Silfverskiöld KL, Andersson CH: Two new
methods of tendon repair: An in vitro evaluation of tensile
strength, and gap formation, J Hand Surg [Am] 18:58–65,
1993, Figure 1b.)

242
Section 2:  Primary Flexor Tendon Surgery
points (Figure 21-2B) without an epitendon suture,
showed a 2-mm gap force of about 4 kg and a maximum
strength of about 6 kg. These values were about three
times greater than two-strand core repairs without a
peripheral suture and quite similar in strength to Wade’s
peripheral type repair. Both the Wade and the Savage
studies indicated the very powerful effect of multistrand
repairs whether the sutures be peripherally or centrally
placed. I used my method clinically and without formal
physiotherapy produced excellent outcomes with a low
rupture rate.23 Interestingly, Sandow and McMahon’s
(1996) anchor point in their modified six-strand Savage
repair24 and Xie and Tang’s (2005) type B “embedded
cross-lock” anchor point25 were the same as my type 2
anchor point in the 1985 study.
Other simpler methods of creating six-strand core
repairs have evolved. In 1996, Sandow and McMahon24
published a simpler anchor point (as noted earlier)
(Figure 21-3) with reduced needle passes, now known
as the Adelaide repair in four-strand form, also has a
simple epitendon suture. In 1994, Tang et al26 com­
bined the principles of Savage’s six-strand repair with
Tsuge et al’s (1975)27 two-strand technique, thus creat­
ing a six-strand Tsuge-type repair. A simple peripheral
epitendon suture was added to prevent gapping.
A way of inserting two strands easily emerged with
Gill et al’s (1999)28 use of a looped suture; being a
single needle swaged onto two ends of a nylon suture,
it could be easily anchored into the tendon by a single
transverse pass of the needle through tendon, with the
needle hooking through the loop. Then it was run down
to the cut end and into the other tendon end where the
suture was inserted about 1 cm; one thread was cut and
the other was passed transversely to create an anchor
point and was knotted at the correct length: an epiten­
don suture was added to prevent gapping. Critics will
notice that although there are six suture strands, there
are only two anchor points on each side, so the repair
is potentially weaker than a repair with three anchor
points.
Four-strand core repairs have proliferated, as a com­
promise to the complexity of six-strand repairs, where
smaller tendons do not accommodate six strands and
three anchor points, and for those who fear, with some
Figure 21-2  A six-strand Savage repair. A, Position of
anchor points. B, Six type 3 anchor points: six strands.
(Modified from Savage R: In vitro studies of a new method of
flexor tendon repair, J Hand Surg [Br] 10:135–141, 1985,
Figure 6a, b.)
Anchor
point
Dorsal
Anchor
point
Anchor
point
Tendon
Tendon
1
3
5
2
4
6
B
A
Figure 21-3  A six-strand Sandow single-cross grasp:
modified Savage technique. (Modified from Sandow MJ,
McMahon MM: Single-cross grasp six-strand repair for acute
flexor tenorrhaphy: Modified Savage technique, Atlas Hand
Clin 1:65–76, 1996, Figure 16.)
this is easy to achieve with seven suture bites on each
side of the tendon: their repair was combined with a
two-strand modified Kessler core suture. Suture material
in Silfverskiöld’s cross-stitch was left on the outside of
the tendon, something that has worried many surgeons,
but many recent studies have used the repair without
apparent extra adhesion.
In 1985 I published a laboratory study on the deriva­
tion of a new technique of core suture.22 The principle
was to use the most effective suture material, to find the
most effective junction of suture onto tendon (then
called “grasp” but which I now call “anchor point”), and
to use multiple strands of suture material. In addition,
the three anchor points were positioned on the radial
lateral, ulnar lateral, and volar sectors of the tendon to
avoid the dorsal vascular supply to the tendon (Figure
21-2A). I tested three patterns/types of anchor point:
type 1 (least intricate) had the lowest strength and the
greatest slippage under tension, and type 3, the most
complex anchor point, gripped the tendon fibers reli­
ably with little slippage, such that in the test on pig
extensor tendon, at the point of ultimate failure either
the suture snapped or a piece of tendon was stripped
out from the tendon substance. The type 2 anchor point
characteristics were between those of types 1 and 3. A
six-strand repair created by using the “type 3” anchor

243
Chapter 21:  The Evolution of End-to-End Surgical Tendon Repairs
a proportionate increase in ultimate strength related
to core strand number. Load sharing between suture
strands appears to be an important part of the repair
and it has been argued that a less complex anchor point
contributes to equal load sharing between suture strands
and thus overall improved strength across the whole
repair.
An important element of repair design is that sutures
should pass from the cut end to the anchor point, and
then back to the cut end immediately, then to the
opposite cut end to make another anchor point, and
then return to the cut end, etc., so that the sutures cross­
ing the tendon gap are multiplied (Figure 21-5A),
according to Xie et al.33 In the modified Kessler repair,
the suture crosses from the first to the second anchor
point by crossing to the opposite side of the same
tendon end, losing the mechanical effect of suture cross­
ing the tendon gap, halving the number of strands in
the repair and creating coils to lengthen under tension
(Figure 21-5B).
Studies have demonstrated the very clear mechanical
advantage of increasing the suture caliber from size 4-0
to 3-0 but, again, in smaller tendons, larger sutures
may not sit comfortably within the tendon. More con­
troversial is the choice of suture material, weighing
the advantage of tougher materials (fiberwire, braided
polyester) with a rough surface against more flexible
materials (polyethylene, polypropylene), which are
slightly weaker but have a with a smooth surface. There
justification, that extra damage to the tendon may result
from more needle insertions. Patterns used include
combinations of modified Kessler, in line, at right
angles, and side by side, and looped sutures have been
paired, tripled, or looped at one end, the “M Tang”
described by Wang et al29 in 2003 and Cao and Tang30
in 2005 (Figure 21-4).
Optimum anchor point position is about 1 cm from
the cut end, according to Cao et al (2006)31 and Tang
et al (2005),32 and optimum anchor point diameter is
about 2 mm, according to Xie et al (2005).33 Despite
much research, it is not clear exactly what style of anchor
point is best; indeed, it may be emerging from Xie and
Tang’s (2005) studies25 and Viinikainen et al (2004)34
that perhaps the anchor point pattern is not as impor­
tant as the multiplication of anchor points and strand
numbers. At least this may be true for repairs treated by
the current favored controlled semiactive mobilization
regimen, but it might not be true, for example, if the
repair was not protected after surgery, although this has
not been tried formally.
It is a possibility that anchor point type cannot be
tested adequately in pig flexor tendons, for Hausmann
et al35 and Peltz et al36 showed there was more resistance
to suture pull-out in pig flexors than sheep or human
flexor tendons in 2009 and 2010. Similarly, a single pull
test could be insufficiently sensitive to distinguish one
anchor point type from another and cyclic load tests
would better mimic real life: In 2005, Matheson et al37
showed minor differences and poor survivorship for
both simple and Silfverskiöld peripheral repairs but
marked differences comparing these repairs to the Savage
repair, which showed 100% survivorship under cyclic
tests. In 1998, Thurman et al,38 in cyclic load tests
in human tendons, showed less gapping with four-
strand Strickland repairs and six-strand Savage repairs
compared to two-strand original Kessler repairs and
Figure 21-4  A six-strand Tang method—three Tsuge
sutures and six-strand M-Tang method. (Modified from
Wang B, Xie RG, Tang JB: Biomechanical analysis of a
modification of Tang method of tendon repair, J Hand Surg
[Br] 28:347–350, 2003, Figure 2.)
Tang method
M-Tang method
Figure 21-5  A, Effective anchor points with no slack to
allow lengthening in a four-strand repair. B, Ineffective grip
of tendon by suture allowing lengthening in a two-strand
repair. (Modified from Xie RG, Xue HG, Gu JH, et al: Effects
of locking area on strength of 2- and 4-strand locking
tendon repairs, J Hand Surg [Am] 30:455–460, 2005,
Figure 4.)
F
F
F
F
F
F
A
B

244
Section 2:  Primary Flexor Tendon Surgery
while in 1994, Elliot et al42 reported a 4% rupture rate,
again raising the possibility that the surgical repair was
not sufficiently strong for CAM. Among other possible
differences between the two series, the repair techniques
were slightly different: even though both used a two-
strand core suture together with the simple over-and-
over epitendon suture, the 46% rupture rate was
associated with the probably weaker modified Kessler
core suture popularized by Kleinert, and the 4% rupture
rate was associated with a more original version of
Kessler’s, which Elliot describes as Kessler/Kirchmayer
(personal communication). The Manchester group now
uses four or six strand core sutures with much improved
results with CAM. There are other examples of this real­
ization around the globe.
SUMMARY
It does appear that 90 years later we have more or less
met Bunnell’s goal. Robust suture repairs have evolved
by multistrand design, both in core sutures and in
peripheral sutures. These are in the main sufficiently
strong to allow early movement but within the restric­
tions of a careful program of exercises and curtailed
activity to protect the repair during healing, and the
results are mainly good. It has taken this time to advance
gradually, learning from use of synthetic low reactivity
sutures, correcting errors in the principles of repair
and mobilization, resolving misunderstandings of the
pathology of healing, and applying simple mechanics
to suture design.
is no advantage to absorbable materials, which may
weaken and cause tissue reaction.
RELATION OF POSTSURGICAL MOBILIZATION
METHODS TO SURGICAL REPAIR TECHNIQUE
Once it became clear that postsurgical mobilization of
fingers could result in excellent outcomes, varying forms
of postsurgical mobilization have evolved and, I believe,
have influenced the way in which surgical repair is
carried out.
In 1973, Kleinert et al, using “dynamic passive mobi­
lization,” showed us that with a relative weak tendon
repair, good mobility could be achieved. However, some
of the poor results with repair rupture39 and repair
stretching and subsequent scar formation40 suggested
that with a stronger tendon repair technique, these poor
outcomes could be improved.
What we now call “controlled active mobilization”
(CAM), initially described by Small et al41 in 1989, dis­
carded the elastic band of Kleinert, replacing gentle
active movement within a tendon splint, and the authors
showed good results in the majority of cases but a 10%
rupture rate. Interestingly, they used the original Kessler/
Mason-Allen core suture with the epitendon part of
Kleinert’s repair technique but not the modified Kessler
core suture, which arguably is weaker.
Following the good results of Small et al, CAM has
been used widely (with variations in exact technique),
although in the United Kingdom at least, it did not yield
the good results in all centers. For example in 1998,
Peck et al,39 in Manchester, reported a 46% rupture rate,
References
1. Bunnell S: Repair of tendons in the fingers, Surg Gynec Obstet
35:88–97, 1922.
2. Lahey FH: A tendon suture which permits immediate motion,
Boston Med Surg J 188:851–852, 1923.
3. Bunnell S: Surgery of the Hand, Philadelphia/London/
Montreal, 1944, JB Lippincott.
4. Mason ML, Allen HS: The rate of healing of tendons: An
experimental study of tensile strength, Ann Surg 113:424–
459, 1941.
5. Kessler I: The “grasping” technique for tendon repair, Hand
5:253–255, 1973.
6. Verdan CE: Primary repair of flexor tendons, J Bone Joint Surg
(Am) 42:647–657, 1960.
7. Kleinert HE, Kutz JE, Atasoy E, et al: Primary repair of flexor
tendons, Orthop Clin North Am 4:865–876, 1973.
8. Kleinert HE, Schepels S, Gill T: Flexor tendon injuries, Surg
Clin North Am 61:267–286, 1981.
9. Mason ML, Shearon CG: The process of tendon repair: An
experimental study of tendon suture and tendon graft, Arch
Surg 25:615–692, 1932.
10. Garlock JH: Repair of wounds of the flexor tendons of the
hand, Ann Surg 83:111–122, 1926.
11. Peacock EE: Biological principles in the healing of long
tendons, Surg Clin North Am 45:461–476, 1965.
12. Potenza AD: Mechanisms of healing of digital flexor tendons,
Hand 1:40–41, 1969.
13. Matthews P, Richards H: Factors in the adherence of flexor
tendon after repair: An experimental study in the rabbit,
J Bone Joint Surg (Br) 58:230–236, 1976.
14. Lundborg G: Experimental flexor tendon healing without
adhesion formation: A new concept of tendon nutrition and
intrinsic healing mechanisms. A preliminary report, Hand
8:235–238, 1976.
15. Matthews P, Richards H: The repair potential of digital flexor
tendons: An experimental study, J Bone Joint Surg (Br)
56:618–625, 1974.
16. Mashadi ZB, Amis AA: The effect of locking loops on the
strength of tendon repair, J Hand Surg (Br) 16:35–39, 1991.
17. Pennington DG: The locking loop tendon suture, Plast Recon­
str Surg 63:648–652, 1979.
18. Wade PJF, Wetherell RG, Amis AA: Flexor tendon repair: Sig­
nificant gain in strength from the Halsted peripheral suture
technique, J Hand Surg (Br) 14:232–235, 1989.
19. Kitsis CK, Wade PJF, Krikler SJ, et al: Controlled active motion
following primary flexor tendon repair: A prospective study
over 9 years, J Hand Surg (Br) 23:344–349, 1998.
20. Silfverskiöld KL, Andersson CH: Two new methods of tendon
repair: An in vitro evaluation of tensile strength and gap for­
mation, J Hand Surg (Am) 18:58–65, 1993.
21. Kubota H, Aoki M, Pruitt DL, et al: Mechanical properties of
various circumferential tendon suture techniques, J Hand Surg
(Br) 21:474–480, 1996.

245
Chapter 21:  The Evolution of End-to-End Surgical Tendon Repairs
34. Viinikainen A, Göransson H, Huovinen K, et al: A com­
parative analysis of the biomechanical behaviour of five
flexor tendon core sutures, J Hand Surg (Br) 29:536–543,
2004.
35. Hausmann JT, Vekszler G, Bijak M, et al: Biomechanical com­
parison of modified Kessler and running suture repair in 3
different animal tendons and in human flexor tendons,
J Hand Surg (Am) 34:93–101, 2009.
36. Peltz T, Haddad R, Savage R, et al: A comparison of human,
porcine and ovine deep flexor tendons. What is the ideal
animal model for in vitro flexor tendon studies? Proceedings
of 15th Congress of the FESSH. June 23–26, 2010. Bucharest,
Romania.
37. Matheson G, Nicklin S, Gianoutsous MP, et al: Comparison
of zone II flexor tendon repairs using an in vitro linear cyclic
testing protocol, Clin Biomech 20:718–722, 2005.
38. Thurman RT, Trumble TE, Hanel DP, et al: Two-, four- and
six-strand zone II flexor tendon repairs: an in situ biome­
chanical comparison using a cadaver model, J Hand Surg
(Am) 23:261–265, 1998.
39. Peck FH, Bucher CA, Watson JS, et al: A comparative study of
two methods of controlled mobilization of flexor tendon
repairs in zone 2, J Hand Surg (Br) 23:41–45, 1998.
40. Ejeskar A: Finger flexion force and hand grip strength after
tendon repair, J Hand Surg (Am) 7:61–65, 1982.
41. Small JO, Brennen MD, Colville J: Early active mobilisation
following flexor tendon repair in zone 2, J Hand Surg (Br)
14:383–391, 1989.
42. Elliot D, Moiemen NS, Flemming AF, et al: The rupture
rate of acute flexor tendon repairs mobilized by the con­
trolled active motion regimen, J Hand Surg (Br) 19:607–612,
1994.
22. Savage R: In vitro studies of a new method of flexor tendon
repair, J Hand Surg (Br) 10:135–141, 1985.
23. Savage R, Risitano G: Flexor tendon repair using a “six-strand”
method of repair and early active mobilisation, J Hand Surg
(Br) 14:396–399, 1989.
24. Sandow MJ, McMahon MM: Single-cross grasp six-strand
repair for acute flexor tenorrhaphy: Modified Savage tech­
nique, Atlas Hand Clin 1:41–64, 1996.
25. Xie RG, Tang JB: Investigation of locking configurations for
tendon repair, J Hand Surg (Am) 30:461–465, 2005.
26. Tang JB, Shi D, Gu YQ, et al: Double and multiple looped
suture tendon repair, J Hand Surg (Br) 19:699–703, 1994.
27. Tsuge K, Ikuta Y, Matshuishi Y: Intra-tendinous tendon suture
in the hand: A new technique, Hand 7:250–255, 1975.
28. Gill RS, Lim BH, Shatford RA, et al: A comparative analysis
of the six strand double loop flexor tendon repair and three
other techniques: A human cadaveric study, J Hand Surg (Am)
24:1315–1322, 1999.
29. Wang B, Xie RG, Tang JB: Biomechanical analysis of a modi­
fication of Tang method of tendon repair, J Hand Surg (Br)
28:347–350, 2003.
30. Cao Y, Tang JB: Biomechanical evaluation of a four-strand
modification of the Tang method of tendon repair, J Hand
Surg (Br) 30:374–378, 2005.
31. Cao Y, Zhu B, Xie RG, et al: Influence of core suture purchase
length on strength of four-strand tendon repairs, J Hand Surg
(Am) 31:107–112, 2006.
32. Tang JB, Zhang Y, Cao Y, et al: Core suture purchase affects
strength of tendon repairs, J Hand Surg (Am) 30:1262–1266,
2005.
33. Xie RG, Xue HG, Gu JH, et al: Effects of locking area on
strength of 2- and 4-strand locking tendon repairs, J Hand
Surg (Am) 30:455–460, 2005.

246
CHAPTER
22 
OUTCOMES OF FLEXOR
TENDON REPAIRS AND
METHODS OF EVALUATION
Jin Bo Tang, MD
topics in hand surgery. The reports of outcomes of
primary repair of the injured flexor tendons were spo­
radic before the 1980s, because this practice had not
become the mainstay of surgical repair of the injury in
the early and middle of the last century. From the 1980s
up to the recent years of this century, we have seen a
great number of reports1-31 that document the outcomes
of primary, or delayed primary, tendon repair, end-to-
end surgical tendon repair methods and evolving post­
surgical rehabilitation methods.
OUTCOMES OVER THE PAST 20 YEARS
I reviewed the outcomes of primary tendon repair in
a 2005 article in Hand Clinics.1 An updated review of
the outcomes of flexor tendon repairs in fingers and
thumb over the past 20 years (1989–2009) is provided
in Table 22-1.
A few important observations can be summarized
from an analysis of outcomes over this period (as well
as the results presented in Chapter 13):
1. Good or excellent recovery of function from the
centers involved (which are among the leading
hand surgery units treating flexor tendon inju­
ries) were in around three-fourths, or 75% of the
cases.
2. The most common method of quantifying out­
comes is the original Strickland and Glogovac
method.
3. Primary or delayed primary tendon repairs were
prioritized over secondary repair by all surgeons.
4. Early tendon motion, either passive or active (or
combined), was advocated in all cases by sur­
geons and participating therapists, except for
children.
5. Early active motion regimens were adopted in
the late 1980s, initially producing a high rate of
repair rupture (around 10%) with conventional
(weaker) two-strand repairs, which has been a
great concern.
6. There has been a gradual but clear shift from
using two-strand core suture repairs to stronger
OUTLINE
Reports of clinical flexor tendon repairs over the past
20 years have documented good or excellent recovery
of function in three-fourths of patients treated in the
finest hand centers worldwide. The most significant
changes in conception and technique regarding flexor
tendon repairs are (1) use of stronger surgical repair
techniques (either multistrand core suture or stronger
peripheral suture) and (2) inception and development
of active finger flexion exercise regimens. Over the past
two decades, rupture rates of the flexor tendon repair
in the fingers gradually declined from around 10% to
less than 2% to 4%, and rupture rates of thumb flexor
tendon repair declined from 10% to 17% to 0%. The
original Strickland and Glogovac criteria are most
popularly used to evaluate the recovery of total active
motion of the fingers. A few novel evaluation methods
emerged, including Moiemen-Elliot’s distal interpha­
langeal joint–only method for zone 1 repair outcomes,
and the author’s method, which incorporates modifi­
cations in grading the total active motion, measure­
ment of grip strength, and assessments regarding the
nature of finger motion. In the final part of this chapter,
I discuss a few issues relating to functional evaluation,
such as a framework of comprehensive assessment of
outcomes, the need for separate evaluation of recovery
of the tendon repair in the little fingers, and grading
the expertise of physicians in clinical reports regarding
outcomes.
The well-recognized importance and challenges in flexor
tendon repair in the hand are reflected in the number
of the past investigations (publications) relating to this
subject. The inaugural issue of the widely read Hand
Clinics was devoted to flexor tendon repair. Twenty years
later, in the preface of the second issue of Hand Clinics
for flexor tendon repair in 2005, Daniel Mass and Craig
Phillips wrote that this controversial complex topic “has
produced more articles in the peer-reviewed hand litera­
ture than any other single topic.” It is true that flexor
tendon repairs have been and remain one of the hottest

247
Chapter 22:  Outcomes of Flexor Tendon Repairs and Methods of Evaluation
Continued
Table 22-1  Some Major Reports of Function of Primary Flexor Tendon Repairs Over the Past 20 Years
Years
Authors
Number of
Digits
Zones
Repair Methods
Excellent/
Good*
Rupture
Rate
Finger flexor tendons
1989
Small et al
138
2
2-strand Kessler
77% (TAM)
9.4%
1989
Cullen et al
38
2
2-strand Kessler
78%
6.4%
1989
Savage and Risitano
23
2
6-strand Savage
69% (Buck-
Gramcko)
4%
1989
Pribaz et al
43
2
Becker
70% (White)
7%
1992
Tang and Shi
54
2
2-, 4-, and
6-strand repairs
81%
…
1994
O’Connell et al
95 (children)
1, 2
2-strand methods
69%†
0%
1994
Silfverskiöld and May
55
2
Silfverskiöld
90%†
3.7%
1994
Grobbelaar and
Hudson
38 (children)
1–5
2-strand Kessler
82% (Lister)
7.9%
1994
Elliot et al
244
1, 2
2-strand Kessler
79%
5.8%
1994
Tang et al
51
2
4- or 6-strand
Tang
77% (White)
4%
1996
Sandow and McMahon
23
2
4-strand Savage
78%
0%
1998
Kitsis et al
208
1–5
2-strand Kessler
92%
2.9%
87
2
88%
5.7%
1999
Fitoussi et al
58 (children)
1–5
2-strand Kessler
89%
0%
1999
Harris et al
626
1, 2
2-strand Kessler
…
4.3%
129
1
…
5.0%
397
2
…
4.0%
2006
Elhassan et al
16 (children)
1
Bunnell pull-out
89%†
0%
25 (children)
2
2- and 4-strand
Kessler
71%†
0%
2008
Caulfield et al
416
1–4
4-strand
Strickland
74%
2%
2008
Haffmann et al
51
2
6-strand Lim/Tsai
78%
2%
26
2
2-strand Kessler
43%
11%
2008
Novali and Rouhani
16 (children)
2
6-strand
Strickland
94%
0%
16 (children)
2
2-strand Kessler
88%
6%
Thumb FPL tendon
1989
Percival and Sykes
51
1–3
2-strand Kessler
53% (White)
8%
1992
Nunlev et al
38
1, 2
2-strand Tajima or
2-strand Kessler
…‡
3%
*The results were evaluated with Strickland and Glogovac criteria unless specified.
†These percentages are percent return of TAM judged by Strickland and Glogovac criteria.
‡Excellent and good rates were not reported. The active IP joint motion was, on average, 35°.

248
Section 2:  Primary Flexor Tendon Surgery
treated over a 3-year period. Ninety-eight patients were
followed and graded using the total active range of
motion (TAM) method of the American Society for
Surgery of the Hand (ASSH). The active range of motion
was graded excellent or good in 77% of the digits, fair
in 14%, and poor in 9%. Repair rupture occurred in
9.4% of the fingers (11 digits). The ruptured tendons
were re-repaired immediately and a similar early motion
program was applied. Cullen et al3 treated 34 adult
patients with 70 zone 2 flexor tendon injuries in 38
fingers; 78% fingers were rated excellent or good by the
original Strickland criteria after a mean follow-up of 10
months. Two tendons ruptured during active finger
flexion exercise. Savage and Risitano4 used their six-
strand method of repairs to treat flexor tendon lacera­
tions in 36 fingers followed by protective active finger
flexion exercise; 63% of lacerations were zone 2 and
27% were zone 1, and 69% and 100% (respectively)
achieved an excellent or good result using Buck-
Gramcko’s assessment method.
Silfverskiöld and May5 reported outcomes of use of
an innovative strong peripheral suture (cross-stitch epi­
tendinous suture) combined with a modified Kessler
core suture in treatment of flexor tendon injuries in
zone 2 in 55 digits among 46 patients. For the first 4
weeks after surgery, fingers were mobilized with a com­
bination of active extension and passive and active
flexion. Repair ruptures occurred in two tendons. Six
months after surgery, the mean TAM of the distal and
proximal interphalangeal (DIP and PIP, respectively)
four- or six-strand surgical repairs, or to use
stronger peripheral repair since 1990s.
7. Use of stronger surgical repairs has lowered
rupture rates, but in most reports, ruptures have
not been eliminated, with 2% to 4% rupture rates
frequently recorded.
8. Over the past decade, more surgeons have
adopted (and reported) some type of early active
motion regimen, together with use of stronger
surgical tendon repairs.
9. New tendon repair materials (such as fiberwire)
emerged, and surgeons have reported good func­
tional outcomes and fewer repair ruptures with
fiberwire repair.
10. Multiple methods are currently used to evaluate
and record the outcomes of tendon repair. Some
worldwide consensus on methods of recording
outcomes is necessary to facilitate comparisons
of results.
MAJOR REPORTS AND
IMPORTANT INFORMATION
Over the past 20 years, a number of excellent clinical
reports have highlighted some important facts regarding
primary flexor tendon repairs. In 1989, The Journal of
Hand Surgery (British Volume) simultaneously published
clinical reports of primary tendon repairs with con­
trolled early active tendon motion by Small et al,2 Cullen
et al,3 and Savage and Risitano.4 Small et al2 presented
114 patients with 138 zone 2 flexor tendon injuries
Table 22-1  Some Major Reports of Function of Primary Flexor Tendon Repairs Over the Past 20 Years—cont’d
Years
Authors
Number of
Digits
Zones
Repair Methods
Excellent/
Good*
Rupture
Rate
1999
Sirotakova and Elliot
30 (1st period)
1, 2
2-strand Kessler
70% (White)
17%
73% (Buck-
Gramcko)
39 (2nd period)
1, 2
2-strand Kessler
67% (White)
15%
72% (Buck-
Gramcko)
49 (3rd period)
1, 2
2-strand Kessler
76% (White)
8%
80% (Buck-
Gramcko)
2004
Sirotakova and Elliot
48
1, 2
Silfverskiöld
73% (White)
0%
77% (Buck-
Gramcko)
2009
Giesen and Elliot et al
50
1, 2
6-strand Tang
78% (White)
0%
82% (Buck-
Gramcko)

249
Chapter 22:  Outcomes of Flexor Tendon Repairs and Methods of Evaluation
tendon repairs with looped suture to repair the tendon
injuries in zone 2 with good functional outcomes.
Several other reports also provide interesting and
important information about tendon repairs. Kitsis
et al16 reported results of treatment of 339 flexor tendon
repaired with two-strand modified Kessler core suture
and a Halsted peripheral suture in zones 1 to 4 of 208
fingers. The important message from this large case
series is that among 6 ruptures of the repaired tendons,
5 occurred in zone 2 and one in zone 5. Repair rupture
was most frequent in zone 2 and the rupture rate was
5.7%. Harris et al18 reviewed the results of 728 primary
zone 1 and zone 2 flexor two-strand modified Kessler
core suture tendon repairs in 526 fingers of 440
patients. A total of 23 fingers (6 in zone 1 and 17 in
zone 2) ruptured 28 tendon repairs. One hundred
twenty-nine fingers with zone 1 injuries had a rupture
rate of 5% (6 fingers) and 397 fingers with zone 2 inju­
ries had a rupture rate of 4% (17 fingers). The ruptures
occurred in 6 tendons within the first week, 5 in the
second week, 6 in the third week, 5 in the fourth week,
3 in the fifth week, and 1 in the tenth week. An impor­
tant message from this case series is that rupture can
occur anytime from the first week to the tenth week
after surgery. The “dangerous period” is postsurgical
weeks 1 to 6, and postoperative week 2 is not associ­
ated with a higher incidence of repair ruptures than
week 1 or weeks 3 to 6. More recently, Dowd et al19
analyzed the results of immediate re-repair of zones 1
and 2 flexor tendon repairs that rupture during postop­
erative early motion exercise. The outcomes of the
re-repair of the ruptured tendons were generally poorer
than the primary repairs, but function of re-repaired
tendons were still clinically acceptable, with 9 (24%)
excellent, 10 (27%) good, 5 (14%) fair, and 13 (35%)
poor when assessed by the Strickland and Glogovac
criteria out of a total of 37 fingers with re-repair of the
tendon.
In 2008, The Journal of Hand Surgery (European Volume)
published a series of reports on flexor tendon repairs
using stronger core tendon sutures (multistrand core
sutures) combined with early active flexion exercise.20-22
These reports indicate lower rupture rates among
tendons with strong core tendon repairs but also illus­
trate that it is not always possible to avoid repair rup­
tures. Caulfield et al20 reported 416 tendon repairs in
zones 1 to 4 repaired with a four-strand Strickland core
suture in 272 patients. The results were 74% good or
excellent graded by the Strickland criteria, with only 2%
repair ruptures. They had identical outcomes using
absorbable and nonabsorbable sutures. Hoffmann
et al21 reported 51 fingers of 46 patients undergoing
zone 2 flexor tendon repair using the Lim/Tsai repair
method combined with a Kleinert and Duran early
motion regimen and place-and-hold exercises. Repair
rupture occurred in 1 (2%) of 51 repaired fingers. Two
joints in the remaining fingers was 63° and 94°, respec­
tively. Elliot et al11 reported 233 patients with complete
divisions of the flexor tendons in zones 1 and 2. Their
cases included 203 patients with 224 finger injuries (317
divided flexor tendons) and 20 patients with 30 com­
plete divisions of the flexor pollicis longus (FPL) tendon.
The patients were treated with a controlled active motion
regimen postoperatively. Repaired tendons ruptured in
13 (5.8%) fingers and 5 (16.6%) thumbs. In follow-up
of the patients treated during the last year of the study,
10 (62.5%) of 16 fingers with zone 1 repairs and 50
(79.4%) of the 63 fingers with zone 2 repairs were
rated good or excellent using the original Strickland and
Glogovac criteria.
Emphasis on the need and application of strong
(four- or six-strand) core repairs in clinical tendon
repairs appeared first in Savage and Risitano’s report4 in
the late 1980s, followed by the report of Tang and Shi7
and Tang et al7,12 in 1992 and 1994 and then a series of
reports in Atlas of the Hand Clinics by Taras et al,13
Sandow and McMahon,14 and Lim and Tsai15 in 1996.
In 1992, Tang and Shi7 reported the results of treatment
of 72 flexor tendon injuries in zone 2 in primary
or delayed primary stages; 80.4% of the fingers were
rated good or excellent results on assessment by the
Strickland and Glogovac criteria. Among 72 tendons, 32
tendons were repaired with four- or six-strand core
sutures using looped sutures, together with simple
peripheral stitches. In 1994, Tang et al12 reported 51
fingers from 46 patients with zone 2 flexor tendon lac­
erations. Doubled threads of the looped suture were
placed to repair injured flexor digitorum profundus
(FDP) or superficialis (FDS) tendons, or three threads
of the looped suture to repair the FDP tendons. The
results were good or excellent in 76.5% using White’s
criteria. They reported two repair ruptures (4%) during
the postoperative motion program. Taras et al13 used
double-grasping Kessler-type core and cross-stitch peri­
pheral sutures in repairing 21 flexor tendons (three
FPL, four FDP in zone 1, and 14 FDS or FDP in zone
2) of 14 digits. The patients underwent active finger
flexion motion initiated on the first postoperative day,
including place-and-hold exercise three times weekly
under supervision. Between therapy sessions, a standard
elastic-thread traction passive flexion and active exten­
sion program was maintained. Overall recovery of
digital motion was graded as excellent in 12 and good
in 2. The seven fingers with FDP and FDS repairs in zone
2 averaged 83% recovery of motion. Sandow and
McMahon14 reported 37 FDP tendons in zones 1 to 5
using a modified single-cross six-strand repair based on
the original Savage method followed by early active
tendon motion. Of 23 zone 2 tendon injuries in 18
patients, 78% were rated as good or excellent using the
Strickland and Glogovac criteria, with no ruptures or
need of secondary surgery. Lim and Tsai15 used six-strand

250
Section 2:  Primary Flexor Tendon Surgery
The use of stronger surgical repairs improved out­
comes compared with the earlier reports using weaker
repair methods but has not entirely eliminated postop­
erative repair rupture.20-22 It is the combination of strong
surgical repair, proper venting of the pulleys, and early
active motion, which reported recently,31,39,41 that has
produced more consistently good or excellent outcomes
without postoperative repair ruptures.
The above reports indicate that clinical flexor tendon
repairs had excellent or good functional return in about
three-quarters of primary tendon repairs. Nevertheless,
it should be noted that most of these reports came from
the finest hand surgery centers and that these teams
were supervised by at least one expert hand surgeon
with experience in treating flexor tendon injuries. We
may reasonably assume that outcomes in a general hos­
pital setting might achieve a lower level of success. In
other words, flexor tendon repairs might have been
unsatisfactory in a larger proportion of patients when
treated in a general hospital setting.
Over the past 20 years, the conceptual evolution in
tendon repair has been remarkable and the technical
innovations have been numerous. The most significant
changes are (1) use of stronger surgical repairs (includ­
ing core and peripheral sutures), (2) incorporation of a
variety of early active motion regimens in postsurgical
exercise, and (3) a redefinition of the management of
the pulleys (see Chapters 9 and 10). Compared to those
in earlier reports, the incidence of repair ruptures in
the most recent series has declined, typically being less
than 2% to 4%. Initiation of early active motion in the
late 1980s, use of stronger surgical repair methods and
active motion regimens in the 1990s, and incorporation
of strong surgical repairs and venting the critical pulleys
with delicate early active motion in the past decade
represent three steps in the pursuit of optimal tendon
repair and ideal recovery of function. These efforts have
transformed the outcomes of flexor tendon repairs from
being very unpredictable to their current state: fairly
predictable. My feeling is that when treated by ade­
quately trained hand surgeons, using updated concepts
and thorough knowledge of flexor tendon repairs, the
outcomes of zone 2 repairs are likely to be comparable
with repairs in the other areas of the hand.
METHODS OF EVALUATIONS
The most common method of recording outcomes is
the original Strickland and Glogovac method intro­
duced in 1980 (Table 22-2).32 This grading system is a
simplification of the method postulated by the ASSH
(Table 22-3).33 The Strickland criteria include TAM of
only the DIP and PIP joints, without inclusion of the
metacarpophalangeal (MCP) joint motion as in the
ASSH method. Strickland pointed out that inclusion of
the MCP joint in assessment of zone 2 repairs artificially
biased the results to be better as the MCP joint was
(4%) fingers required tenolysis. In the cases they treated
with the two-strand modified Kessler repair and the
Kleinert and Duran motion regimen alone, they had
repair ruptures in three fingers (11%) and tendon adhe­
sions, or dehiscence, in three fingers (11%), which
required secondary surgery. The good or excellent
outcome rate was 78% with the Lim/Tsai method and
43% with the two-strand Kessler method. Navali and
Rouhani22 reported on 32 flexor tendon repairs in zone
2 of 29 children using either the two-strand modified
Kessler method (16 tendons) or the four-strand Strick­
land method (16 tendons). They achieved good or
excellent outcomes in all fingers, except one rupture and
one fair outcome among the tendons with the two-
strand repair and one fair outcome in the tendons with
the four-strand repair.
In children, repairs of flexor tendons produced gener­
ally good or excellent results. Early motion exercise is
not essential for a tendon to regain a good range of
active motion. These observations were validated in
reports by Elhassan et al23 in 2006 and by Navali and
Rouhani22 in 2008.
Repair of the FPL tendons is usually discussed sepa­
rately from flexor tendon repair in the fingers.24-31 Siro­
takova and Elliot27,30 analyzed the results of primary
repairs of the FPL tendon followed by early active
motion with only the thumb splinted. The first 30
patients were repaired with a Kessler suture and simple
epitendinous suture. The later 49 patients underwent
repair with a Kessler suture and a reinforced epitendi­
nous suture but in a splint with the thumb position
altered and the fingers also splinted. Other reports
include those from Percival and Sykes,24 Noonan and
Blair,25 Nunley et al,26 Fitoussi et al,28 and Kasashima
et al.29 Most recently, Giesen et al31 published the latest
results of repair of the FPL tendon. They analyzed 50
FPL tendon repairs using a six-strand core repair method,
without peripheral sutures. Excellent or good results
were recorded in 78% and 82% of cases (White and
Buck-Gramcko criteria, respectively). No patients rup­
tured repairs as a result of early active mobilization.
None developed tendon dehiscence postoperatively.
The important message from this report is perhaps that
a strong core suture can adequately tolerate early active
tendon motion without the incorporation of peripheral
sutures, which is a conceptual evolution in the treat­
ment of flexor tendon injuries and may lead to signifi­
cant simplification of surgery.
The case series presented by Elliot and his col­
leagues27,30,31 over the past 20 years reflects some unique
findings relating to repair of FPL injury. Their outcomes
were increasingly improved from a rupture rate of 17%
to the present 0% rate. Their methods of tendon repair
and rehabilitation evolved greatly. The thumb flexor
tendon appears capable of returning to near-ideal func­
tion after proper repair and rehabilitation.

251
Chapter 22:  Outcomes of Flexor Tendon Repairs and Methods of Evaluation
requires the injured fingers to recover a normal active
range of motion. In reality, excellent functional status
requires a sufficiently ample TAM, but it does not neces­
sarily demand that the active range of motion returns
to normal. In addition, flexion of the MCP joint does
not relate only to function of the FDS and FDP tendons;
exclusion of the MCP joint from measurement yields
more accurate documentation of TAM of the PIP and
DIP joints. Strickland modified the original criteria in
1985,34 but the modified criteria have not gained popu­
larity and are considered very lenient.
The Buck-Gramcko method consists of TAM (the sum
of angles formed by the MCP, PIP, and DIP joints in
maximal active flexion minus active flexion deficit at
each joint) and the distance between fingertip and distal
palmar crease (Table 22-4).35 This method is mainly
used by members of the German-speaking societies for
surgery of the hand. The Tubiana method is used mostly
always nearly normal.32 Most subsequent authors
adopted Strickland criteria but have not actually stated
why they preferred the Strickland criteria to the ASSH
method. However, my experience indicates that the
ASSH method is very strict in its “excellent” grade, which
Table 22-2  Criteria of Assessment of Functional Outcomes of Flexor Tendon Repairs
% Return of Motion*
Grip Strength†
Quality of Motion‡
Function Grade
Strickland criteria (1980)
85–100 (>150°)
70–84 (125°-149°)
Excellent
50–69 (90°-124°)
Good
0-49 (<90°)
Fair
Moiemen-Elliot criteria (2000)—for
zone 1 injuries, the DIP joint only
Poor
85–100 (>62°)
Excellent
70–84 (51°-61°)
Good
50–69 (37°-50°)
Fair
0–49 (<36°)
Poor
Tang criteria (2007)
90–100
+
Excellent or good
Excellent +
−
Poor
Excellent −
70–89
+
Excellent or good
Good +
−
Poor
Good −
50–69
Fair
30–49
Poor
<30
Failure
*Percent return of a theoretical normal value or contralateral hand. Strickland and Tang criteria use the sum of active range of motion of the
DIP and PIP joints. The theoretical normal is 175°. Moiemen-Elliot criteria use motion of the DIP joint only compared with 74°.
†Grip strength is recorded as (+) when it is greater than that of the contralateral hand (the nondominant hand) or over 70% of that of the
contralateral hand (dominant hand). Otherwise, grip strength is considered abnormal and recorded as (−).
‡Quality of motion is rated on a basis of direct observation of finger motion by the evaluators. It is recorded as “Excellent” when all three
aspects—motion arc, coordination and speed—are normal; as “good” when any two are normal; as “poor” when only one, or none, is
normal.
Function is graded as excellent − or good − when either the grip strength is (−) or quality of motion is “poor.”
Table 22-3  ASSH Method of Evaluation
Grade
TAM*
Excellent
100%, Normal
Good
>75% of the normal side
Fiar
>50% of the normal side
Poor
<50% of the normal side
Worse
worse than before surgery
*Sum of the active range motion of the MCP, PIP, and DIP joints.

252
Section 2:  Primary Flexor Tendon Surgery
Elliot38 in 2000, and my suggestion39 in 2007 of a
method that includes modified criteria to redefine the
grades of TAM of the DIP and PIP joints and incorpora­
tion of grip strength and quality of finger motion into
the criteria (see Table 22-2).
Admittedly, my suggested method initially appears
more complex. However, one can easily break it down
into components that can be used separately. For
example, the simplest method is to measure TAM of the
DIP and PIP joints, and then use the redefined grades
to record outcomes, the essential function of the criteria.
The results can be classified into “excellent,” “good,”
“fair,” “poor,” and “failure,” without a tag of “+” or “−.”
It is noteworthy that I have set a more stringent bar for
grading the results as “excellent,” requiring recovery of
TAM to 90% of the contralateral side. Additionally, I
added the grade “failure” to designate cases that defi­
nitely require further surgery to restore essential digital
function. Nevertheless, the active range of motion
required for grading as good or excellent (i.e., overall
good and excellent rate) remains the same as in the
Strickland method. In this way, reports regarding total
good or excellent rates using the new method are made
comparable to those in the past using the Strickland and
Glogovac method.
By adding measurement of grip strength and assess­
ment of quality of finger motion, the functional status
of digits can be further indicated with “+” or “−” to
provide supplementary information about hand func­
tion. The information relates to fine functional status of
the repaired tendon and indicates the degree of perfec­
tion of hand function attained.
FACTORS AFFECTING OUTCOMES
Adhesion Formation
Adhesion is a frequent cause of incomplete recovery of
active range of digital motion after tendon repairs
(Figure 22-1). Adhesions arise because trauma to the
tendon surface, inappropriate surgical maneuvers, and
lack of sufficient innate healing capacity of the tendon.
The amount and density of the adhesions are decided
by the severity of the trauma to the tendon and periten­
dinous tissues and the originating tissues. The influ­
ences of adhesions on tendon function vary according
to both the amount and density of adhesions. Repair of
clean-cut tendon injuries, when sufficiently moved after
surgery, may not develop adhesion or only develop
filmy adhesions. The filmy adhesions do not affect
tendon gliding. The formation of restrictive adhesions
are seen after extensive trauma to the tendon or its sur­
rounding structures and jeopardize the gliding of
repaired tendons. During postoperative tendon motion,
loose adhesions can be disrupted, or modified, to avoid
reducing the amplitude of tendon motion. However,
moderate or dense adhesions are difficult to alter and
in France. Among the less popular methods in current
use are White’s method (Table 22-5)36 and the linear
measurement method introduced by Boyes and then
Lister et al (Table 22-6).37
Two more recent methods introduced to record out­
comes include the “DIP motion-only” method for the
zone 1 tendon repairs advocated by Moiemen and
Table 22-4  Buck-Gramcko Method
Item
Measurement
Points
Distance finger pulp to
distal crease
0–2.5 cm/>200°
6
Composite flexion
2.5–4 cm/>180°
4
4–6 cm/>150°
2
>6 cm/<150°
0
Extension deficit
0°-30°
3
31°–50°
2
51°–70°
1
>70°
0
TAM*
>160°
6
>140°
4
>120°
2
<120°
0
*Sum of the active range of motion of the MCP, PIP, and DIP
joints.
Table 22-5  The White Method
Grade
TAM*
Excellent
70%–100% of the normal side
Good
60%–69% of the normal side
Fair
40%–59% of the normal side
Poor
<40% of the normal side
*Active range of motion of the interphalageal (IP) joint only.
Table 22-6  Finger-to-Palm Distance Method
(Boyes-Lister)
Grade
Finger-to-Palm
Distance* (cm)
Extension Deficit (°)
I
0–1
0–15
II
1–1.5
16–30
III
1.5–3
31–50
IV
>3
>50
*Distance between the finger pulp and the distal palmar crease in
full finger flexion.

253
Chapter 22:  Outcomes of Flexor Tendon Repairs and Methods of Evaluation
of narrow pulleys. Annular pulleys, particularly the
A2 and A4, are narrow and restrict tendon gliding.
Edematous, or bulky, tendons are easily entrapped
by these pulleys. Edematous and bulky tendons
can trigger on the edge of the sheath or pulley
openings as well. The narrow part of the annular
pulley constricts the gliding tendons and increases
the resistance to tendon movement, which lead to
gapping or elongation of the tendon repair site.
Catching of the tendon on the edges of narrow
pulleys, halting finger flexion or extension, and
causing patients to feel a sudden increase in resis­
tance to finger motion, is a major cause of rupture
of tendon repairs. After surgery, a forceful pull to
overcome the resistance frequently leads to rupture
of the repairs.
4. Unexpected finger motion. During the period of
wearing protective splints or casts, patients may
sustain unexpected finger actions, such as when
falling down on the outstretched hand or sud­
denly gripping. These actions impose a sudden
increase in the force transmitting through the
repaired tendons with a higher risk of ruptures.
5. Misuses of the fingers. Misuses of the repaired
fingers, such as using the hand to lift a heavy
object, may exceed the repair strength of the
tendon and cause rupture.11,18
6. Unprotected active motion. Unprotected active
flexion of the finger can cause the repaired tendon
to disrupt due to great tension in the tendon.
Joint Stiffness
Stiffness of the DIP and PIP joints is frequently observed
during the rehabilitation after primary flexor tendon
repair. Stiffness of finger joints is troublesome. The
postoperative protective finger position may cause
should, instead, be prevented through careful surgery
and well-supervised postoperative treatments. Tenolysis
is indicated when postoperative treatments cannot sub­
stantially improve tendon motion.
Repair Rupture
Among all the consequences of flexor tendon surgery,
repair ruptures are the worst and are of prime concern
to hand surgeons, because they require secondary oper­
ations. If ruptures occur soon after primary repair, direct
re-suture of the ruptured tendons may be attempted; if
ruptures occur at the late period, a secondary tendon
graft is indicated.40 The following factors trigger the rup­
tures of repaired tendons (see Figure 22-1):
1. Overload of the repaired tendons. Forceful or resisted
active flexion or extension of the fingers may
subject the repaired tendons to a load exceeding
the limit of the tensile resistance of the repairs.
2. Tendon edema or bulky tendons. Edema of the
tendons is inevitable after surgery, although its
severity varies among patients. Severely trauma­
tized wounds, extensive soft tissue injuries, long
duration of surgery, and poor surgical repair
maneuvers all contribute to postsurgical edema.
Edema makes the tendon bulky. In addition,
excessive suture materials also contribute to bulki­
ness. A bulky tendon increases the pressure of the
tendon on the surrounding tissues and its friction
against the sheath and pulleys during tendon
mobilization after surgery. A greater force must be
applied to the finger to move the bulky tendons
within the sheath, increasing the likelihood of
ruptures.
3. Triggering in pulleys or edges of opened sheath and
increased resistance of the repaired tendons by presence
Figure 22-1  Factors that affect the outcomes of the flexor tendon repairs in the hand.
Trauma factors
Surgical factors
Postoperative factors (rehabilitation)
Trauma to tendons
Peritendinous injuries
Surgical skills
Treatment options
Strength of repair methods
Healing strength of tendon
Friction of tendon on sheath
Gap formation, edema
Hand position in splint
Motion range of finger joints
Synergistic motion of joints
Incorporation of particular
protocols for joint stiffness
Joint stiffness
Repair rupture
Adhesion formations
Outcome of tendon repairs

254
Section 2:  Primary Flexor Tendon Surgery
DEVELOPMENT OF COMPREHENSIVE,
UNIVERSALLY ACCEPTABLE CRITERIA
Currently, functional recovery after tendon surgery is
assessed by recording TAM of the digits or distance from
the fingertip to the palm. Grip strength and other factors
relating to tendon function have recently been inte­
grated into new criteria.39 Assessment of motion of a
specific joint relating closely to the repaired tendon,
such as assessment of only the DIP joint after zone 1
tendon repairs, has gained popularity.42 For both the
Strickland and the Moiemen-Elliot methods, issues
remain regarding standardizing the criteria of each
grade in terms of joint angles. Whether to use a fixed
degree of finger motion or the motion range of the
contralateral hand as the baseline has not yet been
agreed. In my experience, contralateral hand motion is
always more reliable than a fixed degree of finger motion
as the baseline for grading. I record angles of the affected
fingers and compute a percentile recovery of TAM com­
pared with the contralateral hand.
There are no universal criteria for documenting the
outcomes of flexor tendon repairs. I suggest that the
authors consider reporting the outcomes at different
levels (i.e., details) (Box 22-1). My preliminary sugges­
tions about such a framework are as follows:
 Level 1. The basic level (essential and relating to
active hand motion only) chiefly includes TAM of
the related joints. Ranges of motion are graded
more stringently. Requirements should be more
rigorous for grading the function as “excellent,”
and a separate grade for failure is necessary. Other
grades of the currently popular Strickland criteria
can be unchanged.
The criteria can be used independently as the
basic level criteria; grip strength and items describ­
ing the nature of motion can be integrated to con­
struct the expanded basic level criteria.
 Level 2. The intermediate level (more comprehen­
sive, but limited to hand function). Criteria incor­
porate TAM, grip strength, other components
reflecting hand motion, sensibility, and assess­
ment of vascular circulation. Multiple aspects of
hand function are reflected at this level.
 Level 3. The advanced, or comprehensive, level (com­
prehensive, relating to entire body function, and
incorporating social environmental impacts). Cur­
rently, the World Health Organization (WHO) is
developing universally adaptable sets of criteria to
reflect the functional status of the hand. This set
of criteria takes into consideration whole body
function, activities, and both environmental and
personal factors. One study has utilized this
concept to evaluate function after flexor tendon
contracture of soft tissue structures of finger joints, par­
ticularly the palmar ligaments of the PIP joints. Because
active finger motion requires both an ample gliding
amplitude of the tendons and a normal passive range
of motion of the joints, to improve the outcomes of
tendon repairs, emphasis should be placed upon pre­
venting stiffness of these joints. Rubber band traction to
the fingertip holds the PIP joint in a flexed position for
long periods, which may lead to development of palmar
ligament contracture and joint stiffness. Modifications
have been suggested to improve the finger motion and
reduce the risk of finger stiffness. In recent years, we
have begun to see fewer cases of finger stiffness after the
use of more delicate active and passive finger motion
regimens, without rubber band traction.20,21,41
Extent of Injuries
Extensive destruction of the peritendinous structures,
the presence of tendon defects, and epitendinous abra­
sion of the tendon are frequently associated with
poorer recovery of function. It is sometimes difficult to
judge whether primary tendon repairs are justified for
wounds that do not involve clean cuts but where direct
approximation of the severed stumps is still possible.
These wounds, which are typified by loss of soft tissues
(sometimes with a short segment of flexor tendons
and a portion of pulleys) over a limited area of the
fingers or palm and defects of soft tissues, are border­
line contraindications to primary tendon repair. Repair
of tendon injuries in severely traumatized wounds
may exacerbate adhesion formation and increases the
likelihood of requiring secondary tenolysis or tendon
reconstruction.
Surgical Skill
Surgical skill affects the outcomes greatly. The flexor
tendon system is composed of anatomic structures
with an intricate biomechanical relationship. Satisfac­
tory repairs of the tendons and associated structures,
particularly those in the intrasynovial regions, remain a
challenge even to an experienced surgeon. In practice,
these difficult injuries are frequently treated by residents
or general orthopedic (or plastic) surgeons without suf­
ficient expertise of flexor tendon surgery. With currently
available knowledge and technical advances, favorable
outcomes may be achieved by an experienced surgeon,
but an individual who lacks expertise may affect repairs
no better than those seen decades ago. It should be
emphasized that when no surgeons experienced in
tendon surgery are readily available, patients should
be referred to hand centers with more experience in
dealing with flexor tendon injuries. Alternatively, after
primary closure of the skin wounds, tendon injuries are
repaired at a delayed primary stage by an experienced
surgeon.

255
Chapter 22:  Outcomes of Flexor Tendon Repairs and Methods of Evaluation
repairs.43 The criteria will reflect entire body func­
tion after flexor tendon repairs.
I expect that, in the future, surgeons and therapists will
record in medical charts, report to social welfare offices,
and publish in the medical literature using criteria with
different details, according to the needs of individual
patients and different levels of academic rigor. Basic
Level criteria may meet the essential requirement. If the
impact of tendon injuries (recovery) on the whole hand
or entire human being is a matter of concern, the record­
ing should move up to Levels 2 and 3.
CONSIDERATIONS RELATING TO OUTCOMES
At the closing of this section on primary flexor tendon
repair, I bring to the attention two issues relating to
outcomes that are likely to receive more attention in the
future.
Repair of the Flexor Tendons in the
Little Fingers
It is now a well-accepted practice to analyze tendon
repairs in the thumb separately from other digits. In
contrast, another issue gaining little specific attention is
repair of the flexor tendons in the little finger. The flexor
tendons in the little finger have a few special character­
istics. The flexor tendons (Figure 22-2) (1) are the
smallest in diameter, (2) move along a multidirection­
ally and significantly curved path, not only flexing but
also curving laterally during finger flexion, (3) are
subject to large structural variations of the sheath and
pulleys, with the A2 pulley being especially narrow, and
(4) have a greater gliding curvature of the tendon, which
predisposes the tendon to a greater bending force.
Repairs in this finger are easier to disrupt.19,44,45 Dowd
et al19 identified a much higher rupture rate and
re-rupture rate and poorer results in the little finger. I
have found it difficult to pass the repaired FDP tendon
under the A2 pulley. Another practical difficulty is that
it is hard to accommodate repairs of over four strands;
although I can put in a six-strand repair, it makes the
tendon gliding surface remarkable rough.
I would bring repair of the flexor tendon injuries in
the little finger to the attention of hand surgeons. I
suggest that this injury be discussed as a separate injury
and its repair outcomes be reported separately, as we do
for tendon repairs in the thumb. The anatomical fea­
tures and biomechanics of the tendon in the little finger
are very different from those of other fingers, and treat­
ment of injuries in the little fingers might also have to
be different from those in the index, middle, and ring
fingers. Unsatisfactory return of function and repair
failure plausibly are more frequent in the little finger.19
Repair of flexor tendon injuries in the little finger
produce less satisfactory function compared with the
other fingers, which remains a concern.
Report Levels of Expertise of Physicians
in Clinical Studies
It is well known and has been popularly written that
outcomes of flexor tendon repairs are expertise depen­
dent. However, currently we do not have any objective
method to scale the expertise of caregivers. Thus far, the
*Include the MCP joint when injuries are proximal to this joint (i.e., zone 3 and proximal); otherwise, exclude the MCP joint.
Level 1—Basic: Relating to Motion of the Finger/Hand Only
My suggested 5 grades according to the criteria:
Excellent: 90% to 100% TAM* of the contralateral side
Good: 70% to 89%
Fair: 50% to 69%
Poor: 30% to 49%
Failure: <30%
Basic expanded: Relating to hand motion and force
Add: Grip strength and quality of finger motion (detailed in Table 22-2)
Level 2—Intermediate: Relating to Hand Motion, Force, Sensation, Circulation, and Cosmetics
Add: Sensation of the hand, blood circulation, and cosmetic appearance
Level 3—Comprehensive: Relating to Hand Function and Cosmetics; Impact on Whole Body; Personal, Social,
and Environmental
Add:
 Whole body function
 Activities
 Environmental and personal factors
Box 22-1
Suggestion About Three Levels in Recording Outcomes of Tendon Repairs in the Hand

256
Section 2:  Primary Flexor Tendon Surgery
A
B
C
Figure 22-2  A, The little finger is much smaller in size
and its flexion arc is much smaller than that of the other
finger. B and C, The pulleys are very narrow and tendons
are much smaller in the little finger, and the repaired FDP
tendon is relatively unsmooth with multi-strand core
suture. The treatment of the tendon injuries in this finger
should be separately discussed; different principles or
methods may be required.
reports, including those of flexor tendon repairs, have
rarely contained information about the degree of profi­
ciency of the caregivers.
This is a topic that does not limit to the reports of
outcomes of flexor tendon repairs. However, flexor
tendon repairs are a perfect example of such needs. To
varying extents in all medical practices, treatment out­
comes are influenced—and biased—by the skill with
which medical personnel perform procedures. In clini­
cal studies, because experience with implemented tech­
niques is not paralleled by job position, simply
categorizing caregivers as residents, attending surgeons,
or consultants, etc., provides little, if any, scientific infor­
mation regarding their expertise in specific techniques
adopted. Individuals grouped into one such category
often differ in their abilities to perform treatment
techniques.
On account of such differences, in reports of treat­
ments that rely heavily on methodology, the use of
technique-performance evaluation criteria is critical for
clear, objective interpretation of treatment outcomes.
This concept holds particular importance in comparable
studies conducted in different institutions or geographi­
cal areas—which exhibit divergences in prevalence of,
and consequently physicians’ exposure to, the study
condition.
To date, no report of clinical outcomes of flexor
tendon repair has documented the level of expertise of
the surgeon(s). We cannot be certain whether worse
outcomes of a specific surgical method relate to the level
of mastery of the surgeon or to the techniques
themselves.
I suggest that expertise levels of the caregivers who
conduct the treatment be reported, perhaps, under
“Methods.” An example of such criteria is detailed in
Table 22-7.46 In reporting the expertise, we should note
that the documented expertise levels are those of the
caregivers performing the treatment, rather than the
expertise of the senior authors of the report. The exper­
tise levels should also relate to specific techniques (or
disorders) under investigation, not to the caregivers’
overall expertise in practice.

257
Chapter 22:  Outcomes of Flexor Tendon Repairs and Methods of Evaluation
References
Table 22-7  Levels and Criteria Suggested to Document Expertise of the Physicians in Reporting
Clinical Outcomes
Level
Category
Criteria
I
Nonspecialist
A physician who is under training, or is a general practitioner of medicine or surgery.
II
Specialist—less
experienced
A physician who has completed training and is a specialist in a subspecialty of
medical or surgical techniques involved, but who has not yet acquired in-depth
knowledge or high-volume experience in the use of the techniques pertinent to the
report.
His or her less degree of experience can be judged by his or her shorter duration of
practice (for example, less than five years) as a specialist, or limited exposure to the
investigated disorder.
III
Specialist—experienced
A physician who has obtained sufficient experience in the use of the techniques
pertinent to the report.
He or she has practiced as a specialist over a longer period (five years or beyond),
with reasonably extensive exposure to the disorder.
IV
Specialist—highly
experienced
A specialist who possesses in-depth knowledge/treatment experience with use of
the techniques being investigated.
This experience is best indicated by having performed, or having been involved as a
leading participant in, scholastic studies relevant to the disorder or techniques being
investigated.
V
Expert
A highly experienced specialist who has made recognized contribution to
advancements of knowledge/treatments related to the disorder being investigated.
Or, a physician who has pioneered the technique in the report.
This category is distinct in that the physician need not first be a specialist if fulfilling
the criteria.
1. Tang JB: Clinical outcomes associated with flexor tendon
repair, Hand Clin 21:199–210, 2005.
2. Small JO, Brennen MD, Colville J: Early active mobilization
following flexor tendon repair in zone 2, J Hand Surg (Br)
14:383–391, 1989.
3. Cullen KW, Tolhurst P, Lang D, et al: Flexor tendon repair in
zone 2 followed by controlled active mobilization, J Hand
Surg (Br) 14:392–395, 1989.
4. Savage R, Risitano G: Flexor tendon repair using a “six strand”
method of repair and early active mobilization, J Hand Surg
(Br) 14:396–399, 1989.
5. Pribaz JJ, Morrison WA, Macleod AM: Primary repair of flexor
tendons in no man’s land using the Becker repair, J Hand Surg
(Br) 14:400–405, 1989.
6. May EJ, Silfverskiöld KL, Sollerman CJ: The correlation
between controlled range of motion with dynamic traction
and results after flexor tendon repair in zone II, J Hand Surg
(Am) 17:1133–1139, 1992.
7. Tang JB, Shi D: Subdivision of flexor tendon “no man’s land”
and different treatment methods in each sub-zone. A prelimi­
nary report, Chin Med J 105:60–68, 1992.
8. O’Connell SJ, Moore MM, Strickland JW, et al: Results of zone
I and zone II flexor tendon repairs in children, J Hand Surg
(Am) 19:48–52, 1994.
9. Silfverskiöld KL, May EJ: Flexor tendon repair in zone II with
a new suture technique and an early mobilization program
combining passive and active flexion, J Hand Surg (Am)
19:53–60, 1994.
10. Grobbelaar AO, Hudson DA: Flexor tendon injuries in chil­
dren, J Hand Surg (Br) 19:696–698, 1994.
11. Elliot D, Moiemen NS, Flemming AFS, et al: The rupture
rate of acute flexor tendon repairs mobilized by the con­
trolled active motion regimen, J Hand Surg (Br) 19:607–612,
1994.
12. Tang JB, Shi D, Gu YQ, et al: Double and multiple looped
suture tendon repair, J Hand Surg (Br) 19:699–703, 1994.
13. Taras JS, Skahen JR, Raphael JS, et al: The double-grasping
and cross-stitch for acute flexor tendon repair, Atlas Hand Clin
1:13–28, 1996.
14. Sandow MJ, McMahon MM: Single-cross grasp six-strand
repair for acute flexor tendon tenorrhaphy, Atlas Hand Clin
1:41–64, 1996.
15. Lim BH, Tsai TM: The six-strand techniques for flexor tendon
repair, Atlas Hand Clin 1:65–76, 1996.
16. Kitsis CK, Wade PJF, Krikler SJ, et al: Controlled active motion
following primary flexor tendon repair: a prospective study
over 9 years, J Hand Surg (Br) 23:344–349, 1998.
17. Yii NW, Urban M, Elliot D: A prospective study of flexor
tendon repair in zone 5, J Hand Surg (Br) 23:642–648, 1998.
18. Harris SB, Harris D, Foster AJ, et al: The aetiology of acute
rupture of flexor tendon repairs in zones 1 and 2 of the fingers
during early mobilization, J Hand Surg (Br) 24:275–280,
1999.
19. Dowd MB, Figus A, Harris SB, et al: The results of immediate
re-repair of zone 1 and 2 primary flexor tendon repairs which
rupture, J Hand Surg (Br) 31:507–513, 2006.
20. Caulfield RH, Maleki-Tabrizi A, Patel H, et al: Comparison of
zones 1 to 4 flexor tendon repairs using absorbable and
unabsorbable four-strand core sutures, J Hand Surg (Eur)
33:412–417, 2008.

258
Section 2:  Primary Flexor Tendon Surgery
33. Kleinert HE, Verdan C: Report of the committee on tendon
injuries, J Hand Surg (Am) 8(Suppl):794–798, 1983.
34. Strickland JW: Results of flexor tendon surgery in zone II,
Hand Clin 1:167–179, 1985.
35. Buck-Gramcko D, Dietrich FE, Gogge S: Evaluation criteria
in follow-up studies of flexor tendon therapy, Handchirurgie
8:65–69, 1976.
36. White WL: Secondary restoration of finger flexion by digital
tendon grafts an evaluation of seventy-six cases, Am J Surg
91:662–668, 1956.
37. Lister GD, Kleinert HE, Kutz JE, et al: Primary flexor tendon
repair followed by immediate controlled mobilization, J
Hand Surg 2:441–451, 1977.
38. Moiemen NS, Elliot D: Primary flexor tendon repair in zone
1, J Hand Surg (Br) 25:78–84, 2000.
39. Tang JB: Indications, methods, postoperative motion and
outcome evaluation of primary flexor tendon repairs in Zone
2, J Hand Surg (Eur) 32:118–129, 2007.
40. Liu TK, Yang RS: Flexor tendon graft for late management of
isolated rupture of the profundus tendon, J Trauma 43:103–
106, 1997.
41. Al-Qattan MM, Al-Turaiki TM: Flexor tendon repair in zone
2 using a six-strand ‘figure of eight’ suture, J Hand Surg (Eur)
34:322–328, 2009.
42. Schaller P, Baer W: Motion-stable flexor tendon repair with
the Mantero technique in the distal part of the fingers, J Hand
Surg (Eur) 35:51–55, 2010.
43. Oltman R, Neises G, Scheible D, et al: ICF components
of corresponding outcome measures in flexor tendon
rehabilitation—A systematic review, BMC Musculoskelet Disord
9:139, 2008.
44. Tang JB, Cao Y, Xie RG: Effects of tension direction on
strength of tendon repair, J Hand Surg (Am) 26:1105–1110,
2001.
45. Tang JB, Xu Y, Wang B: Repair strength of tendons of varying
gliding curvature: A study in a curvilinear model, J Hand Surg
(Am) 28:243–249, 2003.
46. Tang JB: Re: Levels of experience of surgeons in clinical
studies, J Hand Surg (Eur) 34:137–138, 2009.
21. Hoffmann GL, Büchler U, Vögelin E: Clinical results of flexor
tendon repair in zone II using a six-strand double-loop tech­
nique compared with a two-strand technique, J Hand Surg
(Eur) 33:418–423, 2008.
22. Navali AM, Rouhani A: Zone 2 flexor tendon repair in young
children: a comparative study of four-strand versus two-
strand repair, J Hand Surg (Eur) 33:424–429, 2008.
23. Elhassan B, Moran SL, Bravo C, et al: Factors that influence
the outcome of zone I and zone II flexor tendon repairs in
children, J Hand Surg (Am) 31:1661–1666, 2006.
24. Percival NJ, Sykes PJ: Flexor pollicis longus tendon repair: a
comparison between dynamic and static splintage, J Hand
Surg (Br) 14:412–415, 1989.
25. Noonan KJ, Blair WF: Long-term follow-up of primary flexor
pollicis longus tenorrhaphies, J Hand Surg (Am) 16:653–662,
1991.
26. Nunley JA, Levin LS, Devito D, et al: Direct end-to-end repair
of flexor pollicis longus tendon lacerations, J Hand Surg (Am)
17:118–121, 1992.
27. Sirotakova M, Elliot D: Early active mobilization of primary
repairs of the flexor pollicis longus tendon, J Hand Surg (Br)
24:647–653, 1999.
28. Fitoussi F, Mazda K, Frajman JM, et al: Repair of the flexor
pollicis longus tendon in children, J Bone Joint Surg (Br)
82:1177–1180, 2000.
29. Kasashima T, Kato H, Minami A: Factors influencing prognosis
after direct repair of the flexor pollicis longus tendon: multi­
variate regression model analysis, Hand Surg 7:171–176, 2002.
30. Sirotakova M, Elliot D: Early active mobilization of primary
repairs of the flexor pollicis longus tendon with two Kessler
two-strand core sutures and a strengthened circumferential
suture, J Hand Surg (Br) 29:531–535, 2004.
31. Giesen T, Sirotakova M, Copsey AJ, et al: Flexor pollicis
longus primary repair: further experience with the Tang tech­
nique and controlled active mobilisation, J Hand Surg (Eur)
34:758–761, 2009.
32. Strickland JW, Glogovac SV: Digital function following flexor
tendon repair in zone II: a comparison of immobilization
and controlled passive motion techniques, J Hand Surg (Am)
5:537–543, 1980.

259
SECTION
3
SECONDARY FLEXOR
TENDON SURGERY

This page intentionally left blank

261
CHAPTER
23 
TENOLYSIS
Nada Berry, MD, and Peter C. Amadio, MD
insufficient active motion, a cooperative patient, and
access to postoperative therapy. Only when all these
conditions are met should the surgeon consider pro­
ceeding with tenolysis.
A supple digit is important. Releasing contractures
and freeing adhesions at the same time make therapy
unnecessarily complex. Where possible, correcting con­
tractures by serial casting or dynamic splinting should
be performed. Tenolysis must be performed through
supple skin. Skin grafts may be replaced with flaps
beforehand. It is of course logical to require a plateau
in therapy prior to tenolysis; if the patient is still making
progress nonoperatively, an intervention might reverse
rather than accelerate progress. A cooperative patient is
also essential. A patient who, for example, is not willing
to manage their own therapy every day for 6 weeks or
longer, defer vacations and other personal gratification,
and otherwise demonstrate a strong desire to assume
responsibility for their own recovery is unlikely to
benefit from tenolysis. Finally, patients need to have
access to therapy supervision postoperatively. A patient
who lives at a long distance from the surgeon and thera­
pist, or who is prevented from access due to financial or
insurance constraints, will similarly fail to achieve the
maximum benefit from tenolysis surgery.
TIMING OF SURGERY
Tenolysis should be carried out after therapy has been
exhausted and the patient does not show improvement
despite conscientious effort. All scars and affected sur­
rounding tissues should be supple, without evidence of
inflammation. Scar erythema and firmness indicate cor­
ollary deep tissue reaction and should advise the surgeon
against additional surgery.
All fractures and wounds should be healed, and
chronic infections cleared. Joint contractures must be
mobilized. Patient-dependent passive joint motion
should be near normal prior to tenolysis.9 Occasionally,
serial casting or a dynamic external fixator may be useful
prior to the procedure to correct stubborn joint contrac­
tures, especially of the proximal interphalangeal joint.10
Some authors have previously recommended waiting
6 months prior to secondary tendon surgery, as strip­
ping of the tissues surrounding the tendon may devas­
cularize the healing tendon scar and can lead to late
tendon rupture.5 Other studies show that delayed
OUTLINE
Tenolysis is often necessary when adhesions develop
and seriously affect tendon movement. Tenolysis is
indicated for a supple finger with insufficient active
motion in a cooperative patient after plateau in therapy
has been reached. Ideally, tenolysis is done in an
awake patient under local anesthesia, so that active
motion can be assessed continuously. Special knives
may help free the tendons from beneath intact pulleys.
Any patient being consented for tenolysis should also
be consented for tendon grafting, since it is impossible
to be sure preoperatively that adequate tendon integ­
rity will be present postoperatively. Most commonly,
two-stage grafting is appropriate to salvage a failed
attempt at tenolysis. Early active motion is critical to
the success of tenolysis. Special protocols may be nec­
essary if the tendons are noted to be frayed at the
completion of the tenolysis.
Tendon adhesions form frequently after tendon repairs
and reconstructions, phalangeal fractures, and deep
tissue (e.g., flexor sheath) infections.1 Multiple postop­
erative modalities including early active range of motion
protocols have been developed to optimize tendon
gliding. Aggressive therapy may be sufficient to restore
full range of motion for mild adhesions.2 Alternatively,
tenolysis can be a beneficial procedure for patients who
have provided sufficient effort during vigorous therapy
and have plateaued in their range of motion progress.3-8
One must approach these surgeries with caution, as
additional surgery on a less than supremely compliant
patient can lead to further edema, scaring, and worsen­
ing stiffness. Patient and physician expectations should
be thoroughly discussed as it may be difficult to gain
back the hand function, even if the adhesion is not
extensive. Patients with severe trauma requiring multiple
secondary procedures, including nerve repair, tendon
grafting, capsulotomies and osteotomies, those older
than 40 years, tenolysis delayed by a year, and diffuse
adhesions have worse prognosis than those with isolated
tendon injuries and short segments of adhesions.9
INDICATIONS
The indications for tenolysis are a supple digit, a plateau
in motion following a well-supervised therapy program,

262
Section 3:  Secondary Flexor Tendon Surgery
tenolysis after a year results in a decreased postoperative
improvement, possibly due to a developing joint con­
tracture.11 It has been our practice to perform tenolysis
when all of the following conditions have been met: the
digit and soft tissues are supple and well perfused; active
motion is unacceptable to the patient; a plateau in
therapy progress has been reached, with no improve­
ment in motion over at least 4 weeks, and the patient
has been cooperative with the therapy regimen. It is rare
for these conditions to be met in less than 3 or even 4
months from the time of initial tendon injury or repair,
but it is our opinion that tenolysis can be safely per­
formed as soon as 3 months after the injury, providing
that previously mentioned criteria have been met.
SURGICAL TECHNIQUE
Patient involvement during the tenolysis procedure is
considered important by many surgeons and is advis­
able whenever possible. Use of local anesthesia, with or
without limited sedation, allows the patient to aid in
confirmation of release of all the motion-limiting adhe­
sions.7,12-14 Intraoperative active range of motion is the
best predictor of adequate release. In addition, visual­
izing the expected outcome may motivate the patient to
work through the tenderness of the fresh incision and
new edema. While a sterile forearm tourniquet may be
better tolerated than an upper arm tourniquet during
the procedure, it is our preference to eschew all tourni­
quet use, in favor of the “wide-awake” technique advo­
cated by Lalonde and others.13,15
If “wide-awake” surgery is not feasible, a traction test
can be used.16 The involved tendon is exposed proximal
to the area of injury and retracted until digit flexion is
visualized (Figure 23-1). Restrictions in movement
indicate incomplete adhesion release. However, this test
is not infallible; adhesions between muscle bellies may
be well proximal to the zone of initial injury and surgery,
Figure 23-1  Traction test is performed on a patient after
tenolysis. Complete finger flexion is noted in this patient
under general anesthesia, indicating adequate tenolysis. She
had previously undergone flexor tendon repair in zone 2.
Figure 23-2  Extensor tenolysis is performed through an
additional incision in this patient with a healed proximal
phalanx fracture and limited active range of digital motion.
Figure 23-3  Tenolysis knives are very useful in releasing
dense adhesions between the tendon and the flexor sheath.
Meals tenolysis knives shown are available from George
Tiemann & Co, Hauppauge, NY.
especially in longer standing cases. Only “wide awake”
surgery can detect such adhesions.
The incision for tenolysis is usually placed over the
previous scar and extended proximally and distally so
that the tendon and other vital structures can be
approached from virgin tissue. If there is a choice, a
mid-lateral incision may have an advantage of providing
easy access to both flexors and extensors, such as in cases
of fractures or known extensor tendon injury (Figure
23-2). This approach is also useful when resistant joint
contractures need to be released.17
In addition to the preservation of the crucial A2 and
A4 pulleys, the flexor tendon sheath should be mini­
mally traumatized. Releasing any of the pulleys exposes
the tendon to the surrounding tissues and predisposes
it to further scarring. In addition, active grip applies
additional force to the remaining pulleys, leading to
potential pulley rupture and bow stringing of the tendon.
A transverse opening is made into the tendon sheath.
Tenotomes (Figure 23-3) or Freer elevator can be passed

263
Chapter 23:  Tenolysis
immediately, than to have this happen in the postopera­
tive period.
In some cases the profundus is intact but the
superficialis is not. In such cases, the superficialis can
be resected, always with care not to compromise the
blood supply or integrity of the profundus tendon
or the pulley status. A similar approach can be taken
if the superficialis tendon is intact and the profundus
is not.
We prefer early active range of motion and aggressive
postoperative therapy to prevent further adhesion for­
mation. New studies involving absorbable mechanical
barriers and lubricants to prevent adhesion formation
and substances to improve tendon gliding show pro­
mising results both experimentally22-24 and clinically.25
Antimetabolites such as 5-fluorouracil may also be of
benefit.26 We tend to use these adjunctive agents on a
second tenolysis but not usually with initial procedures,
because usually the result of a first tenolysis is satisfac­
tory without these agents, which may compromise
wound healing.
REHABILITATION
Early active motion is a key factor in the success of any
tenolysis procedure and should be instituted as soon as
possible.27-32 If hemostasis is not a concern and the pain
is tolerable, we prefer to begin rehabilitation in the
operating room, taking advantage of the “wide-awake”
method, and continue in the recovery room. In general,
we prefer light dressings and regular active motions—if
under the pulley system to separate the tendon from the
sheath. Active finger flexion can aid in lysis of distal
adhesions. If the critical A2 pulley is not intact, it can
be reconstructed, but this is likely to compromise the
result of the tenolysis, since vigorous active postopera­
tive motion will be contraindicated in the face of a
pulley reconstruction.18 In such cases, a staged recon­
struction may be an alternative consideration.19,20 Pre­
operative magnetic resonance imaging (MRI) (Figure
23-4) or ultrasound may aid in determining pulley
status.21 In contrast, the A4 pulley may not be so critical;
if it is not intact and the profundus is also compromised
but the superficialis is intact, then the distal interpha­
langeal joint can be stabilized and the profundus
excised. A superficialis finger can be superior to both a
staged reconstruction and even a heroic attempt at
tenolysis of the profundus and A4 reconstruction.
The tendon quality should be examined. Pseudo-
tendon formation in the gap between the tendon
ends can mislead the surgeon into thinking that the
tendon healing is complete and result in surgical failure.
Hunter has advised that a tendon defect involving
more than 30% of the tendon is grounds to consider a
staged tendon reconstruction,20 and we agree with this
approach. Such a defect may be obvious initially or at
the completion of tenolysis. An additional advantage of
the “wide-awake” approach is that the patient can be
asked to make a strong fist at the end of the procedure.
It is far better that a compromised tendon rupture intra­
operatively, where a staged procedure can then be done
Figure 23-4  A, MR image shows normal relationship between the flexor tendon (T), overlying bone, and the A3 pulley. B,
MR image shows increased distance between the tendon and the bone, as well as the attenuated pulley.
A
B
T
A3

264
Section 3:  Secondary Flexor Tendon Surgery
of dehiscence, it is important to use a very careful tech­
nique when elevating skin flaps, in ensuring hemostasis
at the time of closure, and in establishing a secure skin
closure.
RESULTS
Overall, larger reviews in the literature suggest that 80%
of patients show improvement with tenolysis proce­
dures, 10% remain mostly unchanged, and 10% worsen
their range of motion or sustain a rupture of the tenoly­
sis.4,5,14,25 It is unusual for a patient to regain active
control of the full passive range that is present preopera­
tively, however. These statistics and expectations should
be discussed with the patient preoperatively. Tenolysis
requires a combination of patient cooperation, skillful
therapy, and careful surgery to achieve a satisfactory
outcome.
possible, 10 repetitions every hour while awake or at
least two sessions per day. Splinting is usually not neces­
sary unless the tendon is frayed, in which case the wrist
and metacarpophalangeal joints can be splinted in
flexion, especially while the patient sleeps. Formal
therapy continues until the patient has reached a plateau
in recovery, with no improvement in active range of
motion over 3 weeks. In general, patients will maintain
their gains in the long term if they improve during the
first postoperative week and maintain the gains for the
following 2 to 3 weeks.5
Indwelling catheters for administration of local anes­
thetic can be useful postoperatively.33-37 Painful, rigor­
ous activity may lead to additional swelling and an
overall setback.
Wound dehiscence may require a halt in therapy,
especially if the tendon is exposed. To minimize the risk
References
1. Matthews P, Richards H: Factors in the adherence of flexor
tendon after repair: An experimental study in the rabbit,
J Bone Joint Surg (Br) 58:230–236, 1976.
2. Ipsen T, Barfred T: Early mobilization after flexor tendon
grafting for isolated profundus tendon lesions, Scand J Plast
Reconstr Hand Surg 22:163–167, 1988.
3. Azari KK, Meals RA: Flexor tenolysis, Hand Clin 21:211–217,
2005.
4. Eggli S, Dietsche A, Egglis S, et al: Tenolysis after combined
digital injuries in zone II, Ann Plast Surg 55:266–271, 2005.
5. Fetrow KO: Tenolysis in the hand and wrist. A clinical evalu­
ation of two hundred and twenty flexor and extensor tenoly­
ses, J Bone Joint Surg (Am) 49:667–685, 1967.
6. Schneider LH: Tenolysis and capsulectomy after hand frac­
tures, Clin Orthop Relat Res 327:72–78, 1996.
7. Strickland JW: Flexor tenolysis, Hand Clin 1:121–132, 1985.
8. Yamazaki H, Kato H, Uchiyama S, et al: Results of tenolysis
for flexor tendon adhesion after phalangeal fracture, J Hand
Surg (Eur) 33:557–560, 2008.
9. Strickland JW: Flexor tendon surgery. Part 2: Free tendon
grafts and tenolysis, J Hand Surg (Br) 14:368–382, 1989.
10. Kawakatsu M, Ishikawa K, Terai T, et al: Distraction arthrolysis
using an external fixator and flexor tenolysis for proximal
interphalangeal joint extension contracture after severe crush
injury, J Hand Surg (Am) 35:1457–1462, 2010.
11. Verdan C: Tenolysis. In Verdan C, editor: Tendon Surgery of the
Hand, Edinburgh, 1979, Churchill Livingstone, pp 137–142.
12. Hunter JM, Schneider LH, Dumont J, et al: A dynamic
approach to problems of hand function using local anesthe­
sia supplemented by intravenous fentanyl-droperidol, Clin
Orthop Relat Res 104:112–115, 1974.
13. Lalonde DH: Wide-awake flexor tendon repair, Plast Reconstr
Surg 123:623–625, 2009.
14. Whitaker JH, Strickland JW, Ellis RK: The role of flexor teno­
lysis in the palm and digits, J Hand Surg (Am) 2:462–470,
1977.
15. Higgins A, Lalonde DH, Bell M, et al: Avoiding flexor tendon
repair rupture with intraoperative total active movement
examination, Plast Reconstr Surg 126:941–945, 2010.
16. Strickland JW: Flexor tendon injuries. Part 5. Flexor tenolysis,
rehabilitation and results, Orthop Rev 16:137–153, 1987.
17. Saffar P: Total anterior teno-arthrolysis. Report of 72 cases,
Ann Chir Main 2:345–350, 1983.
18. Foucher G, Lenoble E, Ben Youssef K, et al: A post-operative
regime after digital flexor tenolysis. A series of 72 patients,
J Hand Surg (Br) 18:35–40, 1993.
19. Amadio PC, Wood MB, Cooney WP 3rd, et al: Staged flexor
tendon reconstruction in the fingers and hand, J Hand Surg
(Am) 13:559–562, 1988.
20. Hunter JM: Staged flexor tendon reconstruction, J Hand Surg
(Am) 8:789–793, 1983.
21. Drape JL, Silbermann-Hoffman O, Houvet P, et al: Complica­
tions of flexor tendon repair in the hand: MR imaging assess­
ment, Radiology 198:219–224, 1996.
22. Zhao C, Sun YL, Amadio PC, et al: Surface treatment of flexor
tendon autografts with carbodiimide-derivatized hyaluronic
acid. An in vivo canine model, J Bone Joint Surg (Am) 88:2181–
2191, 2006.
23. Zhao C, Sun YL, Kirk RL, et al: Effects of a lubricin-containing
compound on the results of flexor tendon repair in a canine
model in vivo, J Bone Joint Surg (Am) 92:1453–1461, 2010.
24. Karakurum G, Buyukbebeci O, Kalender M, et al: Seprafilm
interposition for preventing adhesion formation after tenoly­
sis. An experimental study on the chicken flexor tendons,
J Surg Res 113:195–200, 2003.
25. Riccio M, Battiston B, Pajardi G, et al: Efficiency of Hyaloglide
in the prevention of the recurrence of adhesions after tenoly­
sis of flexor tendons in zone II: A randomized, controlled,
multicentre clinical trial, J Hand Surg (Eur) 35:130–138,
2010.
26. Zhao C, Zobitz ME, Sun YL, et al: Surface treatment with
5-fluorouracil after flexor tendon repair in a canine in vivo
model, J Bone Joint Surg (Am) 91:2673–2682, 2009.
27. Alba CD, LaStayo P: Postoperative management of function­
ally restrictive muscular adherence, A corollary to surgical
tenolysis: A case report, J Hand Ther 14:43–50, 2001.
28. de Soras X, Thomas D, Guinard D, et al: Use of an implanted
electrode for rehabilitation after tenolysis of the flexor
tendons, Ann Chir Main Memb Super 13:317–327, 1994.
29. Feldscher SB, Schneider LH: Flexor tenolysis, Hand Surg 7:61–
74, 2002.
30. Foucher G, Marin Braun FM: An original mobilization technic
after tenolysis of zone II flexor tendons, Ann Chir Main 8:
252–253, 1989.
31. Goloborod’ko SA: Postoperative management of flexor tenol­
ysis, J Hand Ther 12:330–332, 1999.

265
Chapter 23:  Tenolysis
32. McCarthy JA, Lesker PA, Peterson WW, et al: Continuous
passive motion as an adjunct therapy for tenolysis, J Hand
Surg (Br) 11:88–90, 1986.
33. Braun C, Bauer M, Bühren V: Experiences with continuous
nerve block of the wrist, Handchir Mikrochir Plast Chir 23:
207–209, 1991.
34. Jablecki J, Syrko M: The application of nerve block in early
post-operative rehabilitation after tenolysis of the flexor
tendon, Orthop Traumatol Rehabil 7:646–650, 2005.
35. Kirchhoff R, Jensen PB, Nielsen NS, et al: Repeated digital
nerve block for pain control after tenolysis, Scand J Plast
Reconstr Hand Surg 34:257–258, 2000.
36. Kulkarni M, Elliot D: Local anaesthetic infusion for postop­
erative pain, J Hand Surg (Br) 28:300–306, 2003.
37. Lurf M, Leixnering M: Ultrasound-guided ulnar nerve
catheter placement in the forearm for postoperative pain
relief and physiotherapy, Acta Anaesthesiol Scand 53:261–263,
2009.

266
CHAPTER
24 
SINGLE-STAGE FREE TENDON
GRAFTING FOR FLEXOR
TENDON INJURY IN FINGERS
Pierre-Yves Barthel, MD, and Pierre Mansat, MD, PhD
 Delayed presentation of closed FDP avulsion inju­
ries from the insertion associated with significant
tendon retraction.
 Closed rupture of the flexor tendons at any level
from zone 1 to zone 3, with retraction of the
proximal tendon that does not permit a direct
end-to-end repair.
Preoperative prerequisites include the following:
 The wound should be healed, the joints should be
free of contracture, and maximum passive motion
should have been obtained.
 There is no extensive scarring.
 Passive movements are full, or nearly full.
 The circulation is satisfactory.
 At least one digital nerve in the affected digit is
intact.
Contraindications to single-stage free tendon grafting
are:
 Extensive scarring, pulley incompetence, or joint
contracture.
 Insensate or poorly vascularized digits.
The best indication for single-stage free tendon graft for
flexor tendon injury is the grade 1 hand according to
Boyes’ classification (Table 24-1)1 or according to Merle
and Dautel’s classification (Table 24-2).2 For grade 2 or
3 hands, staged tendon grafting is preferred.
In cases of disruption of both the FDP and FDS
tendons, general surgical principles for this procedure
include:
 Place only one graft in each finger.
 Use a graft of a suitable caliber to fit into the finger.
 Place the proximal junction outside the tendon
sheath.
 Avoid damage to the fingernail or fingertip, in
making the distal junction.
 Ensure adequate graft tension.
In cases with an intact or functioning FDS tendon, the
following additional principles apply:
OUTLINE
Successful primary flexor tendon repair has limited the
indications for flexor tendon grafting. Tendon grafting
is an alternative treatment for patients with neglected
digital flexor tendon lacerations and after failure of
primary flexor tendon repair. Tendon grafting can be
considered if the local conditions of the hand are
favorable. However, the outcomes after flexor tendon
grafting are inferior to those of primary flexor tendon
repair. A decision must be made after careful discus­
sion with the patient of the risks and benefits of this
procedure and specific needs of the patient.
Single-stage free tendon grafting is a common secondary
procedure for tendons not suitable for or not treated by
primary repair. The surgery must be carefully discussed
between the surgeon and the patient to evaluate the risk
and benefits of the procedure. This operation is usually
indicated for injury of both flexor digitorum profundus
(FDP) and flexor digitorum superficialis (FDS) tendons
in zone 2 in the presence of a good pulley system and
without serious scarring. The decision as to the need for
staged reconstruction is often made during surgery
depending on the state of the pulleys and the extent of
adhesions. The injured tendons are excised and replaced
by a harvested tendon graft sutured to the base of the
distal phalanx at the FDP insertion and to a proximal
motor in the palm or distal forearm.
INDICATIONS
Single-stage free tendon graft is carried out as an initial
procedure in selective cases including:
 Flexor tendon injuries with segmental tendon loss.
 Severe damage to peritendinous tissues, or obvious
risk of wound infection that prevented primary
repair.
 Delay in repair of more than 3 weeks that com­
promises primary repair.

267
Chapter 24:  Single-Stage Free Tendon Grafting for Flexor Tendon Injury in Fingers
distal FDP tendon is preserved. The remainder of the
damaged FDP tendon is excised distal to the lumbrical
origin in the palm. The FDS tendon is resected, leaving
1 or 2 cm of the insertion of the FDS tendon to provide
stability and avoid hyperextension deformity of the
interphalangeal joint (Figure 24-1).
Graft
The most frequent donors are the palmaris longus and
the plantaris tendons. The palmaris longus tendon is
preferred for palm-to-fingertip reconstruction, and the
plantaris for forearm-to-fingertip reconstruction. Other
possibilities include the long extensors of the three
middle toes, the toe flexors, the extensor indicis pro­
prius, and the extensor digiti minimi (Figure 24-2).
Placement of the Graft
A suture is placed at the end of the graft, and the graft
is passed atraumatically under the pulleys using a flex­
ible catheter or a Silastic Hunter rod in a distal-to-prox­
imal direction. Care must be taken to minimize trauma
to the intact sheath to limit the risk of adhesion and
secondary stiffness (see Figure 24-1).
Sequence of Attachment of the Graft Ends
Creating the distal juncture first may allow easy judg­
ment of tension and determination of the graft length
in the palm. The proximal suture is accomplished after
the distal juncture has been completed.3 The alternative
is to create the proximal juncture first, with adjustment
of the graft length at the fingertip. In our opinion, the
latter is more difficult and less reliable for restoring
proper tension to the graft.
Distal End Attachment
Several techniques are available to anchor the tendon
graft to the distal phalanx, depending primarily on the
length and condition of the distal FDP tendon. (1) If
enough FDP tendon is available, the graft can be sutured
to this tendon or through it. (2) When the length of
distal FDP tendon is not sufficient, the tendon is attached
to the distal phalanx using a tendon-to-bone pull-out
technique. A drill hole is made in the palmar cortex
of the distal phalanx into which straight needles are
passed through the nail plate. A sterile button is usually
used to tie over the surface on top of the nail plate. This
conventional method has been criticized for damaging
the fingernail and risking infection.4 Although surgeons
still use this method, many have started using other
methods for distal juncture that have no interference
with the nail complex (see Figure 24-1).
More recent alternatives to the conventional method
include (1) drilling a transverse drill hole across the
base of the distal phalanx with the graft passed through
and sutured to itself. This technique allows for easy
judgment of graft tension, but the graft must be thin
 Carefully explain the risk of no or limited recovery
of digital function before surgery.
 FDS-only fingers may have sufficient function
without using free tendon grafting to reconstruct
the FDP tendon.
 Never sacrifice an intact FDS or damage the surface
of the FDS tendon during surgery.
OPERATIVE METHODS
Injuries to Both FDP and FDS Tendons
Incision
Surgery is usually performed under local or regional
anesthesia with tourniquet control. A Bruner-type or
mid-lateral incision is used to expose the tendon system.
The lower limb must be prepared if the surgeon intends
to harvest a graft from there.
Dissection of the Flexor Tendons
The neurovascular bundles are preserved throughout
the dissection. The tendon sheath is explored and pro­
tected as much as possible. Approximately 1 cm of the
Table 24-1  Boyes’ Preoperative Classification1
Grade
Preoperative Condition
1
Good: Minimal scar with mobile joints and no
trophic changes
2
Cicatrix: Heavy skin scarring because of injury
or prior surgery; deep scarring because of failed
primary repair or infection
3
Joint damage: Injury to joint with restricted
range of motion
4
Nerve damage: Injury to digital nerves resulting
in trophic changes in finger
5
Multiple damages: Involvement of multiple
fingers with combination of above problems
Table 24-2  Preoperative Classification Based on
Modified Grading of Merle and Dautel2
Grade
Preoperative Condition
1
Minimal or mild scar without major damages to
digital vascular bundles and nerves
2
Extensive scar in the digit causing or together
with:
 Destruction to multiple annular pulleys
 Joint stiffness
 Damage of one digital neurovascular
bundle
3
Serious scar in the digit with:
 Injury of bilateral digital neurovascular
bundles
 Injury of main vascular supplies to the hand

268
Section 3:  Secondary Flexor Tendon Surgery
complications of pull-out techniques.5-8 Metallic anchors
are preferred to avoid possible complications of auto­
lysis associated with bioabsorbable anchors.9 Surgeons
in some units have virtually abandoned the conven­
tional pull-out suture, replacing it with these metallic
anchors for distal juncture. For patients with poor bone
quality, a pull-out suture method is preferred to avoid
to pass through the bony tunnel. Risk of fracture must
be known (Figure 24-3). (2) For patients with good
bone quality, some authors have proposed distal fixation
with metallic suture anchors (Figure 24-4) and non­
absorbable sutures. Biomechanical studies have shown
that metallic anchors are sufficient to provide stable
fixation of the graft to the distal phalanx, avoiding the
Figure 24-1  A, Incision and methods of insertion of a grafted tendon into preserved tendon sheath. B, Two methods of
distal tendon juncture. C, Pulvertaft method of proximal tendon juncture.
A Preservation of pulleys and passing the graft within the sheath
B Two methods of distal tendon junction
C Pulvertaft repair for proximal junction

269
Chapter 24:  Single-Stage Free Tendon Grafting for Flexor Tendon Injury in Fingers
After using the pull-out suture technique for many
years, we now use metallic anchor sutures to gain imme­
diate strength of the junction and also to avoid the
complications of a pull-out suture, such as nail dys­
trophy, pulp ulceration, and infection. Furthermore,
because of the strong tendon anchor, immediate mobi­
lization of the finger can be allowed with fewer risks of
anchor site rupture. After preparation of the distal
phalanx by removing the remnant of the distal FDP
tendon, one or two mini-anchors are placed into the
distal phalanx. The tendon graft is guided into position
and secured to the distal phalanx with the sutures
carried by the anchor. Additional sutures of 3-0 PDS
ensure the graft stays well coapted to the distal phalanx
and to the surrounding tissues, including the palmar
plate of the DIP joint. Strength of the distal fixation are
assessed by adding a traction force on the graft. A radio­
graphic examination is always performed to verify that
the metallic anchors are well seated in the distal phalanx.
Proximal Juncture Suture
In the palm, the proximal end of the graft is attached to
the FDP tendon just distal to the lumbrical origin. The
end of the motor (proximal) tendon is opened with a
blade, and the graft is then passed into the split. The
graft is passed transversely in different planes according
to the Pulvertaft technique. Then 3-0 PDS sutures are
used to suture the tendons at the interconnection points,
and the distal split is closed over the graft to obtain a
perfect continuity between the motor tendon and the
graft (see Figure 24-1).
migration of the anchor.10,11 (3) Taras and Kaufmann
describe their method of passing the suture around the
distal phalanx and tying it over the nail.3 The sutures are
threaded onto the end of the grafted tendon with straight
needles. The tendon graft is woven through distal stump
of the FDP tendon. The needles are then passed around
both sides of the distal phalanx and through the middle
third of the nail plate, avoiding the germinal nail matrix;
the ideal point to exit through the nail plate should be
3 to 4 mm distal to the lunula and approximately 2 mm
from the midline. The ends of the suture are tied directly
over the nail plate. At least two reinforcing sutures on
each side are used to reinforce the FDP–graft juncture
(Figure 24-5).
Figure 24-2  Methods of harvesting tendons from donor sites in forearm or lower extremity.
A
C
B
Palmaris longus
Extensor
digitorum
longus
Plantaris tendon
Tendon stripper
Achilles tendon
Figure 24-3  Distal tendon juncture method: passing the
tendon through a drill hole in the distal phalanx.
Gauze
Profundus
stump

270
Section 3:  Secondary Flexor Tendon Surgery
Graft Tensioning
Graft tensioning is estimated by the relaxed position of
the fingers with the wrist in the neutral position. Each
finger should fall into semiflexion, slightly less flexed
than its ulnar neighbor and more flexed than its radial
neighbor. The Pulvertaft technique is especially amen­
able to adjusting tension; if the posture of the hand is
satisfactory with one suture in place, the juncture is
completed by adding additional sutures (Figure 24-6).
Postoperative Care
Postoperatively, the hand is immobilized with a static
dorsal blocking splint with the wrist positioned in
neutral to mild flexion, the metacarpophalangeal (MCP)
joints in 45° of flexion, and the splint distal to the MCP
joint in 0° of extension to allow the finger proximal
and distal interphalangeal (PIP and DIP) joints to rest
in the neutral position. The position of splinting is
changed after 3 to 4 weeks to place the wrist and hand
in the functional position. This splint is retained until
6 weeks after surgery. Passive range of motion exercise
can be instituted 2 or 3 days after surgery, but avoid
extension of the MCP joint and the wrist. At 2 weeks,
slight active short-arc digital flexion and extension exer­
cises can be added, with the MCP joint in flexion if
strong distal suturing is obtained. However, if the distal
sutures are less robust because of poor bone quality or
loss of the distal flexor tendon stump, active mobiliza­
tion is delayed until 4 weeks. At 4 weeks, blocked flexion
exercises of the MCP or PIP joint are started. Usually the
splint is discarded at 6 weeks and resistance exercises
can be started. Passive stretching splint can then be used
if flexion contracture remains.
FDP Tendon Disruption with
FDS Tendon Intact
Surgeons have mixed recommendations regarding
surgery for the patient with complete disruption of the
FDP tendon but an intact FDS tendon. FDS-tendon
Figure 24-4  A, The tools for inserting a mini-anchor into
the distal phalanx. The upper, a drill, is used to drill a hole
on the proximal palmar surface of the bone if necessary, and
the lower, the anchor will be inserted into the bone by the
inserter. The mini-anchor, carrying two sets of suture lines,
is attached to the tip of the inserter. B, Drilling a hole on the
palmar surface of the distal phalanx. C, The mini-anchor is
securely placed into the bone. After that, the tendon is
sutured with the suture lines carried by the anchor.
(Courtesy of Jun Tan, MD, and Jin Bo Tang, MD.)
A
B
C
Figure 24-5  Method of distal tendon attachment to the
distal phalanx. The suture is anchored over the nail surface.

271
Chapter 24:  Single-Stage Free Tendon Grafting for Flexor Tendon Injury in Fingers
FDS tendon if the tendon decussation is tight for the
grafted tendon, but we found this rarely necessary, as the
grafted tendon is usually thinner than the FDP tendon,
though the collapse of the pulleys and sheath occurs
after delay of the surgery. Fitting the grafted tendon and
intact FDS into the sheath or pulley may be difficult. In
such cases, surgeons may consider partial removal of
narrow parts of the sheath or pulley, but obviously the
sheath and pulley should be preserved segmentally to
maintain function of the tendons. Avoiding damage to
an intact, fully functioning FDS tendon is an important
principle. Incision or removal of the FDS tendon is pro­
hibited under any conditions. In some cases, one slip of
the FDS tendon can be removed for graft passage and to
allow gliding of the grafted tendon. However, removal
of a slip of FDS tendon can be risky, which the authors
do not recommend. If extensive scarring in the FDS, or
the flexor tendon bed, or pulleys, is found during surgery,
staged tendon grafting is preferable.
OUTCOMES AND COMPLICATIONS—
PROGNOSIS FACTORS
Outcomes
In general, the outcomes after flexor tendon grafting
are inferior to those of primary flexor tendon repair. In
Boyes and Stark’s study,12 reviewing 1000 flexor-tendon
grafting, 607 at the fingers, the preoperative condition
of the digit was found to be the most important factor
affecting the result of flexor-tendon grafting. In good
cases (grade 1), regardless of age, time since surgery,
specific digit, or tendon used as a graft, 23% of the
fingers flexed the pulp to the palmar crease; only 9%
fingers may function very well without surgical recon­
struction of the FDP tendon. There is also a risk to
damage to the FDS function during surgery. The sur­
geons should carefully evaluate the patient and explain
to the patient about the degree of improvement of func­
tion and possible risk of reconstructing the FDP tendon
with a tendon graft in the presence of an intact FDS
tendon.
First, function of the FDS tendon must be examined
thoroughly. When the FDS tendon is fully functional,
caution must be observed before proposing a free tendon
graft to restore distal joint function. Because most useful
motion is maintained when the FDS tendon is fully
functional, patients may have adapted perfectly, even
with an injured or retracted FDP tendon. The problem
for the patient is mostly about the DIP joint position.
If there is no hyperextension of the joint, functional
adaptation is usual. However, if there is hyperextension
of the joint with instability, procedures for stabilization
can be discussed with the patient. Simple yet effective
procedures, like distal tenodesis or arthrodesis of the
DIP joint, can be proposed. Indications for tendon graft­
ing are discussed in selected cases, especially young
patients with mobile joints and a reasonable need for
active DIP joint function. A graft is probably more fre­
quently justified on the ulnar side of the hand than on
the radial side because of the need to obtain a total arc
of flexion to be able to perform power gripping.
The method of tendon grafting is the same as the
technique described earlier for patients with injuries to
both. In performing grafting for such patients, we take
great care in passing the graft through the intact FDS
tendon decussation. The graft can be laid anterior to the
Figure 24-6  The fingers’ cascade for deciding tension placed on the graft in different fingers.
A
B

272
Section 3:  Secondary Flexor Tendon Surgery
Prognosis Factors
In their series, Boyes and Stark12 outlined different prog­
nostic factors—scarring from injury, additional scarring
from inept previous surgery, or failed primary reparative
procedures compromised the results of secondary
tendon grafting; fingers in which joints had been
damaged or in which the IP joints had become stiffened
from neglect did not respond well to tendon grafting,
even though the joints were mobilized well before
surgery. In fingers with minimum scarring and only
one nerve injured, the results were not impaired, but
fingers with both nerves damaged had much less motion.
The level of tendon injury, whether in the proximal,
middle, or distal part of no-man’s land, was not a deter­
mining factor on the result, nor was the time from
injury to operation. Injury of the tendon of more than
one digit in itself was not important. They found that
the condition of the individual digit determined the
outcome for that finger. Patients over 40 years of age
did not regain as much motion from tendon grafting
as did the patients in younger age groups. The palmaris
longus tendon was the best donor tendon, but there was
little difference noted when a good superficialis tendon
was used.
SUMMARY
Single-stage free tendon grafting is a valuable secondary
reconstructive procedure of the injured deep flexor
tendons of the fingers. It is mainly used after injury to
both flexor tendons in zone 2. The preoperative status
achieved this in the scar (grade 2) and nerve injury
groups (grade 4), while in the joint and salvage groups
(grades 3 and 5), none could flex to make the finger
tip touch the distal palmar crease. McClinton et al13
reviewed 100 tendon grafts for isolated FDP injuries
and reported that 55 digits could be flexed to touch
the distal palmar crease postoperatively. Twenty-four
digits were flexed to bring the fingertip within 1.3 cm,
15 between 1.3 and 2.5 cm, and 6 between 2.5 and
5 cm of the distal palmar crease. Using intrasynovial
tendon grafts and early mobilization, Leversedge et al14
reported that recovery of active motion was 64% for
single-stage reconstruction and 55% for multiple-stage
reconstruction. Single digit reconstruction had the best
outcome, with recovery of total active motion of 73%.
Kotwal and Gupta15 reviewed the results of one-stage
free tendon grafting in 240 patients (264 digits) and
found excellent results in 18.5% of the cases, good
results in 70.5%, and poor results in 11% (Table 24-3).
Complications
The major complications seen after flexor-tendon graft­
ing are hyperextension at the PIP joint, rupture of the
graft in the palm, rupture at the distal juncture, and
trophic ulcers of the finger pulp. A tendon graft may also
become adherent and fail to move the finger through a
functional range of motion. The digit itself may become
so scarred in a flexed position that extension is limited
and total range of the motion is very small, which requires
tenolysis.12-15
Table 24-3  Results of Single-Stage Free Tendon Grafting for Flexor Tendon Reconstruction in the Hand
Authors
Digits
Type of Graft
Results
Complications
Boyes and
Starks12
607
PL 413
PL+ paratenon 11
FDS 128
Plantaris 21
Toe extensor 29
Others 5
Full flexion obtained in:
  23% in grade 1 fingers
  12% in grade 2 fingers
  10% failures (ppd = 5 cm)
Failure (i.e., ppd > 5 cm):
32 fingers
Secondary rupture:
15 in the palm, 7
at distal insertions
PIP recurvatum 13
Hematoma 6
Ulcer of the pulp 2
McClinton
et al13
100
Isolated FDP injury
PL 80
EDC 10
EIP 4
Toe extensor 4
Plantaris 1
FDS 1
ppd:
0 cm: 55
0-1.3 cm: 24
1.4-2.5 cm: 15
2.6-5 cm: 6
AROM DIP joint: 48° on average
Failure:* 13
Not stated
Leversedge
et al14
4
FDL 2nd toe
Excellent 1, good 1, fair 1, poor 1
64% TAM recovery
Rupture 1
Tenolysis 1
Kotwal and
Gupta15
264
PL
Plantaris
Excellent 18.5%
Good: 70.5%
Poor: 11%
Infection 7
Secondary rupture 3
EDC, Extensor digitorum communis; EIP, extensor index proprius; FDL, flexor digitorum longus; PIP, proximal interphalangeal joint;
DIP, distal interphalangeal joint; AROM, active range of motion; TAM, total active motion; PL, palmaris longus; ppd, pulp–palm distance.
*Failure was defined in this study as loss >20° from preoperative PIP joint flexion or DIP joint flexion less than 20°.

273
Chapter 24:  Single-Stage Free Tendon Grafting for Flexor Tendon Injury in Fingers
after flexor tendon grafting are generally inferior to the
results of primary flexor tendon repair. When only the
FDP tendon has been injured with an intact FDS
tendon, caution must be observed before carrying out a
free tendon graft reconstruction of the FDP tendon,
because the patient may have adapted his/her hand
function. Alternative options, such as distal tenodesis
or arthrodesis of the DIP joint, must be discussed with
the patient.
of the digit is of paramount importance. Secondary
single-stage flexor tendon grafting is only considered
when the wound is healed, the joints are free of con­
tracture with maximum passive motion, there is not
extensive scarring, and the patient has good circulation
and soft tissue coverage of the hand. With rigorous
surgical technique, appropriate tensioning of the graft,
and postoperative therapy, patients can recover their
hand function quite remarkably. However, the results
References
1. Boyes JH: Flexor-tendon grafts in the fingers and thumb: An
evaluation of end results, J Bone Joint Surg (Am) 32:489–499,
1950.
2. Merle M, Dautel G: Chirurgie secondaires des tendons fléchis­
seurs. In Michel M, Gilles D, editors: La Main Traumatique–
Tome 2–Chirurgie Secondaire/Le Poignet Traumatique, Paris,
1995, Masson, pp 55–92.
3. Taras JS, Kaufmann RA: Flexor tendon reconstruction. In
Wolfe SW, Hotchkiss RN, Pederson WC, et al, editors: Green’s
Operative Hand Surgery, ed 6, Philadelphia, 2011, Elsevier, pp
207–238.
4. Kang N, Marsh D, Dewar D: The morbidity of the button-over-
nail technique for zone 1 flexor tendon repairs. Should we still
be using this technique? J Hand Surg (Eur) 33:566–570, 2008.
5. Bonin N, Obert L, Jeunet L, et al: Réinsertion du tendon flé­
chisseur par ancre de suture: etude prospective continue avec
mobilisation active précoce, Chir Main 22:305–311, 2003.
6. Brustein M, Pellegrini J, Choueka J, et al: Bone suture anchors
versus the pullout button for repair of distal profundus
tendon injuries: a comparison of strength in human cadaveric
hands, J Hand Surg (Am) 26:489–496, 2001.
7. McCallister WV, Ambrose HC, Katolik LI, et al: Comparison
of pullout button versus suture anchor for zone I flexor
tendon repair, J Hand Surg (Am) 31:246–251, 2006.
8. Latendresse K, Dona E, Scougall PJ, et al: Cyclic testing
of pullout sutures and Micro-Mitek suture anchors in flexor
digitorum profundus tendon distal fixation, J Hand Surg (Am)
30:471–478, 2005.
9. Galano GJ, Jiang KN, Strauch RJ, et al: Inflammatory response
with osteolysis related to a bioabsorbable anchor in the
finger: A case report, Hand (NY) 5:307–312, 2010.
10. Matsuzaki H, Zaegel MA, Gelberman RH, et al: Effect of
suture material and bone quality on the mechanical proper­
ties of zone I flexor tendon-bone reattachment with bone
anchors, J Hand Surg (Am) 33:709–717, 2008.
11. Giannikas D, Athanaselis E, Matzaroglou C, et al: An unusual
complication of Mitek suture anchor use in primary treat­
ment of flexor digitorum profundus tendon laceration: A case
report, Cases J 14:9319, 2009.
12. Boyes JH, Stark HH: Flexor-tendon grafts in the fingers and
thumb. A study of factors influencing results in 1000 cases,
J Bone Joint Surg (Am) 53:1332–1342, 1971.
13. McClinton MA, Curtis RM, Wilgis EF: One hundred tendon
grafts for isolated flexor digitorum profundus injuries, J Hand
Surg (Am) 7:224–229, 1982.
14. Leversedge FJ, Zelouf D, Williams C, et al: Flexor tendon
grafting to the hand: an assessment of the intrasynovial donor
tendon—A preliminary single-cohort study, J Hand Surg (Am)
25:721–730, 2000.
15. Kotwal PP, Gupta V: Neglected tendon and nerve injuries of
the hand, Clin Orthop Relat Res 431:66–71, 2005.

274
CHAPTER
25 
A HISTORICAL PERSPECTIVE
ON FLEXOR TENDON
RECONSTRUCTION AND
SURGICAL PROCEDURES
Andrew E. Farber, DO, and Daniel P. Mass, MD
paving the way for modern tendon repair and
reconstruction.
DEVELOPMENT OF TENDON
RECONSTRUCTION
Until the 1960s, tendon lacerations in zone 2, or “no
man’s land,” were treated with removal of the tendon
with grafting of new tendons. Sterling Bunnell taught “it
is better to remove the tendons entirely from the finger
and graft in new tendons smooth throughout its
length.”2,3 Based on Bunnell’s classic work in the early
twentieth century, repair of tendon lacerations in zone
2, “no man’s land,” involved removal of the tendon and
grafting new tendons.3 The techniques of single-stage free
tendon grafting were later refined by a number of master
hand surgeons, including Pulvertaft, Graham, Littler,
Boyes, and Stark.5-17 Boyes’ large series of tendon grafts
popularized the procedure in 1950s and 1960s.13,14,17
In 1963, Bassett and Carroll subsequently first
described this type of two-stage flexor tendon recon­
struction using a silicone implant.18 In 1965, Hunter
first published his experience with tendon implants for
tendon reconstruction.19 As is widely reported, Hunter
further refined this process in the early 1970s, resulting
in the naming of the Hunter silicone rod and staged
tendon reconstruction techniques currently used.19,20 An
alternative method was described by Paneva-Holevich
in 1969,21 who sutured the proximal cut end of the
flexor digitorum superficialis (FDS) tendon to the proxi­
mal cut end of the profundus tendon in the palm. At
the second stage, the FDS tendon was severed as far
proximal as possible and this end was brought out to
be inserted at the distal phalanx as a pedicle graft. In
1982, Paneva-Holevich reported secondary repair of
324 flexor tendon injuries using pedicle FDS tendon
grafting. In 39 fingers, this technique was combined
with silicone rod implantation at the first stage to
prepare a smooth bed for the graft.22
OUTLINE
Flexor tendon reconstruction is a spectrum of
challenging treatment options ranging from non-
operative treatment to tenolysis to single-stage
and multistage reconstructive surgical procedures. Sec­
ondary tendon reconstruction is indicated in the
patients whose tendon injuries are not treated at the
primary or delayed primary stages, whose injuries
to the tendon are too serious, or where destruction of
the flexor pulleys are too extensive, which prevents
primary repair of the tendon. In the presence of
severe scar in the tendon gliding bed or extensive
pulley damages, both patient and surgeon are to be
prepared for multiple surgical procedures. This chapter
starts with a concise review of the development of
secondary flexor tendon reconstruction, followed by
descriptions of surgical procedures and postoperative
rehabilitation of tendon grafting and staged tendon
reconstructions.
In the tenth century, Avicenna, an Arabian surgeon, was
credited with performing the first tendon repair surgery.
In Europe, however, Galen teachings resulted in infre­
quent tendon repair. Galen did not differentiate between
nerves and tendons, noting that nerves entered muscles
and muscles ended with whitish cords. Accordingly,
teaching at that time (circa 150 AD) involved an intimate
relationship between tendons and nerves, so much so
that physicians feared severe consequences from damag­
ing or even touching a nerve or a tendon. They further
believed that suturing of a nerve or a tendon would
result in pain, convulsions, and gangrene.1
In 1752, Galen’s concept that complications are asso­
ciated with tendon damage was finally refuted by the
work of Albrecht von Haller,1 who demonstrated the
insensibility of tendons. In England around 1850, Syme
reported success with several cases of tendon repair

275
Chapter 25:  A Historical Perspective on Flexor Tendon Reconstruction and Surgical Procedures
critical to maximize the chance for a successful outcome.
Boyes and others have described indications for single-
stage grafting including: (1) injuries with segmental
tendon loss and (2) delayed presentation resulting in the
inability to perform a primary end-to-end repair. This
may include injuries older than 3 to 6 weeks.4,11-17
The hand and finger should also demonstrate at least
one intact digital nerve in the affected finger, a well-
healed wound without extensive scarring, adequate cir­
culation; and near full-passive motion of all joints.
Insufficient passive range of motion should be addressed
with a course of hand therapy prior to considering surgi­
cal intervention. Patients should also be prepared to
participate in a rigorous and complex postoperative
rehabilitation program. This requirement generally
excludes growing children and some elderly patients.19
Excellent surgical technique is imperative and should
be carried out under loupe magnification and tourni­
quet control.
Except in the case of preexisting scar, choice of surgi­
cal incision is generally a matter of surgeon preference.
Options include the Bruner zigzag or mid-lateral inci­
sions. During dissection, neurovascular structures are
identified and protected. Damaged sheath is excised
and effort is made to protect the sheath and pulley
system.23-26 Missing pulleys, in particular the A2 and A4
pulleys, need to be reconstructed.
While several donor tendon options exist, the ipsilat­
eral palmaris longus or plantaris tendons remain the
most common choices. Since these tendons are both
nonsynovial tendons, some surgeons prefer to use toe
extensors from beneath the ankle retinaculum within
the sheath. Toe flexors are also intrasynovial; however,
their sheaths tend to be much shorter than those of
the fingers. In the event of harvesting the tendon from
the lower extremity, the incision in the hand is at
this point covered with moistened sponges and the
tourniquet deflated. Additionally, although vascularized
tendon grafts have been described and investigated, they
are rarely used and little is written about their use in the
English literature.
The FDP insertion site is next identified distally. In
the case of an intact distal FDP, the stump is reflected
and the graft is secured in the method of Bunnell. Spe­
cifically, a 3-0 Prolene or similar suture is placed at the
end of the graft and criss-crossed twice. In an adequate
FDP stump, the graft may be woven to reinforce fixation.
Otherwise, the FDP stump is split and the suture ends
are threaded onto a Keith needle. The needles are then
passed around both sides of the distal phalanx or, alter­
natively, drilled obliquely from the proximal aspect of
the volar distal phalanx, exiting through the middle-
third of the nail plate, approximately 3 to 4 mm distal
to the lunula and approximately 2 mm from the midline.
Whenever possible, avoid letting them exit through the
germinal nail matrix. The sutures are then passed
SURGICAL PROCEDURES
It is helpful to think of flexor tendon reconstruction as
a spectrum of challenging treatment options ranging
from nonoperative treatment to tenolysis to single-stage
and multistage reconstructive surgical procedures. Both
patient and surgeon need to be prepared for multiple
surgical procedures and strict compliance with postop­
erative rehabilitation protocols. It is therefore impera­
tive to carefully consider surgical and nonsurgical
indications when selecting the best treatment protocol.
Additionally, it is essential to discuss the range of out­
comes with the patient including the likelihood of stiff­
ness and the possible need for amputation. In some
cases, doing nothing or performing a primary amputa­
tion or arthrodesis may avoid a series of painful proce­
dures that may ultimately provide or restore minimal
function.
Bunnell’s extensive work on flexor tendon injury and
reconstruction resulted in the widely accepted belief
that injured tendons should be removed and replaced
with new tendon graft.2,3 Bassett, Carroll, and Hunter
refined this recommendation with their works on the
silicone rod, which was designed to create a smooth bed
for tendon grafting. These developments led to the
widespread acceptance and use of the two-stage tendon
reconstruction.
Over the four decades that followed, however, signifi­
cant advancements in flexor tendon repair and recon­
struction led to revised indications for both primary
repair and secondary reconstruction. When considering
secondary repair, the decision needs to be made whether
to graft and reconstruct the tendons in a single stage or
to perform a two-stage repair. This decision requires
careful preoperative and intraoperative examinations.
Regardless of the surgical technique chosen, patients
should be counseled and should be prepared to partici­
pate in a complex rehabilitation program. This may
preclude performing the procedure on children younger
than 3 years of age.
Single-Stage Reconstruction
Preoperatively, patients considered for single-stage
reconstruction with tendon grafting must display a well-
healed, neurovascularly intact digit without excessive
scarring. Intraoperatively, an intact, smooth tendon bed
should be present with an adequate pulley system.
Patients with inadequate soft tissue coverage, those
with significant, dense scarring, or an inadequate pulley
system generally benefit from a two-stage reconstruc­
tion. Additional indications and limitations are described
next.
Although primary flexor tendon repair has become
increasingly popular, select patients can still benefit from
single-stage flexor tendon reconstruction.4-10 Careful
patient selection and intraoperative evaluation are

276
Section 3:  Secondary Flexor Tendon Surgery
extensive destruction of flexor pulleys, those with crush­
ing injuries with extensive soft tissue damage or under­
lying fractures, and patients with extensive scarring of
the flexor tendon bed. Adequate passive range of digital
motion and pliable skin are essential for the patients as
candidates of the staged reconstruction (Figures 25-3
and 25-4).
Hunter, among others, is credited with describing
and refining these techniques with the goal of creating
a new flexor tendon bed allowing for gliding of the
implanted tendon graft. As mentioned earlier, it is
imperative to maintain or recreate a sufficient and func­
tional pulley system. Traditional teaching has said that
at a minimum, the A2 and A4 pulleys at the proximal
and middle phalangeal levels respectively must be in
place. However, more recent data suggest that a four-
pulley system may be superior whenever possible.4 At
the same time, however, tight pulleys can inhibit tendon
graft gliding; accordingly, constricted pulleys can be
dilated. Damaged but intact pulleys should be repaired
Figure 25-2  Position of the fingers and tension set during
operation.
Figure 25-3  Hand demonstrating adequate passive range
of motion.
Figure 25-1  Demonstration of a suture anchor for distal
fixation of the grafted tendon.
through the holes and tied over a padded button.4 If
available, a suture anchor can be used for fixation to the
distal phalanx (Figure 25-1).
The proximal aspect of the graft is then secured to the
FDP tendon just distal to the lumbrical origin. Any
remaining scar or frayed tissue is removed from the
native FDP tendon and a Brand tendon braider or a fine
scalpel is used to pass the graft. The graft should be
passed through several planes in the native tendon and
the tendon junctures are sutured in place.
Estimating proper tension is often difficult, especially
in the anesthetized patient; however, this remains
among the most important aspects of this procedure.
Placing the wrist in a neutral position allows for evalu­
ation of the relaxed position of the fingers.24,27-30 Each
semiflexed finger should be slightly more flexed than
the next radial finger and slightly less flexed than the
next ulnar finger (Figure 25-2). The Pulvertaft weave
technique is commonly used for securing the tendon
graft to the proximal tendon stump in the palm or distal
forearm. This technique is particularly useful in that it
allows for adjustment of graft tension.
Postoperative protocols typically included 3 weeks of
immobilization. More recently, however, active motion
protocols have become prevalent in primary flexor
tendon repairs,31,32
Two-Stage Reconstruction
Despite significant improvements in primary repair
and single-stage tendon reconstruction, two-stage recon­
struction remains a valuable option for some patients,
including patients who suffer severe trauma, those with

277
Chapter 25:  A Historical Perspective on Flexor Tendon Reconstruction and Surgical Procedures
avoid damage to the rod and talc-free gloves should be
used as talc contamination is known to cause implant
adherence.
A tendon passer is used to pass the implant from
proximal to distal or from distal to proximal (Figure
25-6).37-39 The implant is then secured distally below the
FDP stump, taking care to pass sutures through the
Dacron tape in the implant since the silicone itself offers
little holding strength. Hunter rods are also available
fitted with a metal eyelet that can accommodate a
2.0-mm screw for fixation into the distal phalanx. When
using this technique, it is important to avoid penetrating
the dorsal cortex, inuring the nail bed. Additionally, the
screw must be distal enough in the distal phalanx to
avoid blocking DIP motion.
Next, range of motion is performed and observed to
ensure an adequate pulley system and free passage of
the graft with no bowstringing. Revision or additional
pulley reconstructions may be indicated at this point in
the procedure.
Wounds are then irrigated and closed. A bulky dress­
ing is applied, supporting the wrist in approximately
35° of flexion, metacarpophalangeal joints in 60° to
70° of flexion, and the interphalangeal joints relaxed in
extension. Passive motion may be started 2 to 3 days
postoperatively.
To allow for adequate wound healing and develop­
ment of a gliding tendon sheath in response to the
implant, typically 3 months is allowed between stages
of reconstruction. The criteria for proceeding with the
and missing or irreparable pulleys must be reconstructed
(Figure 25-5). A variety of options exists for pulley
reconstruction, including grafting of extensor retinacu­
lum, remnant of free tendon, or synthetic materials.33,34
Bunnell originally recommended reconstructing the
pulley with placement of the tendon graft deep to the
extensor mechanism at the A2 level and superficial at
the A4 level;35 however, Taras and others more recently
suggested placing the tendon graft deep at all levels.36
As in other flexor tendon procedures, volar Bruner
incisions are ideal and should incorporate previous
scars where possible. Flexor tendons are then excised
leaving a 1-cm FDP stump attached to the distal phalanx.
A 1-cm FDS stump should also be maintained at the
middle phalanx. If necessary, collateral ligaments and
the volar plate can be released freeing remaining joint
contractures.
With a sufficient pulley system in place, the silicone
or Hunter rod is placed. A smooth forceps is used to
Figure 25-4  Hand demonstrating pliable skin.
Figure 25-5  Reconstruction of flexor pulley at the middle
of the proximal phalanx.
Figure 25-6  At the first stage, with a Bruner incision in
the middle of the finger and two incisions at the fingertip
and the palm, the Hunter rod is placed to replace the FDP
tendon and to stimulate formation of tendon gliding bed.

278
Section 3:  Secondary Flexor Tendon Surgery
The graft is harvested, the silicone implant is detached
from its distal insertion site, and the graft is sutured to
the proximal end of the implant for passage through the
sheath. The graft is passed from proximal to distal as the
implant is removed.
As previously described, the graft is secured distally.
The incisions are washed and closed and a short-arm
dorsal blocking splint is applied. Typically, the wrist is
placed in neutral, the metacarpophalangeal joints in
45° of flexion and the interphalangeal joints in neutral.
Protected passive range of motion with early controlled
active motion is employed postoperatively.39-41
second stage of reconstruction are similar to those
mentioned for the first stage and include a soft scar, near
full passive range of motion, and intact protective
sensation.
In the second stage of flexor tendon reconstruction,
the implant is removed and a tendon graft is placed.
Previous surgical scars are opened and distally, the
implant is identified and removed from the FDP stump.
Depending on the site of injury, the proximal incision
in the hand or forearm is reopened and the implant is
carefully exposed to avoid injuring the tendon sheath.
Typically, the adjacent profundus motor unit is selected.4
References
21. Paneva-Holevich E: Two-stage tenoplasty in injury of the flexor
tendons of the hand, J Bone Joint Surg (Am) 51:21–32, 1969.
22. Paneva-Holevich E: Two-stage reconstruction of the flexor
tendons, Int Orthop 6:133–138, 1982.
23. Wilson RL: Flexor tendon grafting. Flexor tendon surgery,
Hand Clin 1:97–107, 1985.
24. Watson AB: Some remarks on the repair of flexor tendons in
the hand, with particular reference to the technique of free
grafting, Br J Surg 43:35–42, 1955.
25. Bunnell S: Repair of tendons in the fingers, Surg Gynecol Obstet
35:88–97, 1922.
26. Strickland JW: Delayed treatment of flexor tendon injuries
including grafting, Hand Clin 21:219–243, 2005.
27. Peljovich A, Ratner JA, Marino J: Update of the physiology
and biomechanics of tendon transfer surgery, J Hand Surg
(Am) 35:1365–1369, 2010.
28. Kim SH: A loop-tendon suture for tendon transfer or graft
surgery, J Hand Surg (Am) 32:367–372, 2007.
29. Boyes JH: Operative technique of digital flexor tendon grafts,
Instr Course Lect 10:263–268, 1953.
30. Littler JW: Free tendon grafts in secondary flexor tendon
repair, Am J Surg 74:315–321, 1947.
31. Stenstrom SJ: Functional determination of the flexor tendon
graft length, Plast Reconstr Surg 43:633–634, 1969.
32. Strauch B, de Moura W: Digital flexor tendon sheath: an
anatomic study, J Hand Surg (Am) 10:785–789, 1985.
33. Freilich AM, Chhabra AB. Secondary flexor tendon recon­
struction, a review, J Hand Surg (Am) 32:1436–1442, 2007.
34. Lin GT, Amadio PC, An KN, et al: Biomechanical analysis of
finger flexor pulley reconstruction, J Hand Surg (Br) 14:278–
282, 1989.
35. Bunnell S: Surgery of the Hand, Philadelphia, 1944, JB
Lippincott.
36. Taras JS, Kaufmann RA: Flexor tendon injury. B. Flexor tendon
reconstruction. In Wolfe SW, Hotchkiss RN, Pederson WC,
et al, editors: Green’s Operative Hand Surgery, ed 6, Philadel­
phia, 2011, Churchill Livingstone-Elsevier, pp 207–238.
37. Seradge H, Homan ES, Spiegel PG: Tendon passer, Clin Orthop
Relat Res 155:307–308, 1981.
38. Sourmelis SG, McGrouther DA: Retrieval of the retracted
flexor tendon, J Hand Surg (Br) 12:109–111, 1987.
39. Hunter JM, Blackmore S, Callahan AD: Flexor tendon salvage
using the Hunter tendon implant, J Hand Ther 2:107–113,
1989.
40. Mackin EJ: Physical therapy and the staged tendon graft: pre­
operative and postoperative management. In AAOS Sympo­
sium on Tendon Surgery in the Hand, St Louis, 1975, CV Mosby,
pp 283–291.
41. Stanley BG: Flexor tendon injuries: late solution. Therapist’s
management, Hand Clin 2:139–147, 1986.
1. Adamson JE, Wilson JN: The history of flexor tendon grafting,
J Bone Joint Surg (Am) 43:709–716, 1961.
2. Strickland JW: Development of flexor tendon surgery: twenty-
five years of progress, J Hand Surg (Am) 25:214–235, 2000.
3. Bunnell S: Repair of tendons in the fingers and description
of two new instruments, Surg Gynecol Obstret 26:103–110,
1918.
4. Pulvertaft RG: Indications for tendon grafting. In Hunter JM,
Schneider LH, Mackin EJ, et al, editors: Rehabilitation of the
Hand, St Louis, 1984, CV Mosby, pp 277–279.
5. Colville J: Tendon graft function, Hand 5:152–154, 1973.
6. Mason ML, Allen HS: The rate of healing of tendons, Ann Surg
113:424–459, 1941.
7. Pulvertaft RG: Problems of flexor tendon surgery of the hand,
J Bone Joint Surg (Am) 47:123–132, 1965.
8. Pulvertaft RG: Indications for tendon grafting. In AAOS Sym­
posium on Tendon Surgery in the Hand, St Louis, 1975, CV
Mosby, pp 123.
9. Pulvertaft RG: Indications for tendon grafting. In Hunter JM,
Schneider LH, Mackin EJ, et al, editors: Rehabilitation of the
Hand, St Louis, 1984, CV Mosby, pp 277–279.
10. Schneider LH: Treatment of isolated flexor digitorum profun­
dus injuries by tendon grafting. In Hunter JM, Schneider LH,
Mackin EJ, editors: Flexor Tendon Surgery in the Hand, St Louis,
1986, CV Mosby, pp 518–525.
11. Adamson JE, Wilson JN: The history of flexor tendon grafting,
J Bone Joint Surg (Am) 43:709–716, 1961.
12. Allen HS: Flexor tendon grafting to the hand, Arch Surg
63:362–369, 1951.
13. Boyes JH: Evaluation of results of digital flexor tendon graft,
Am J Surg 89:1116–1119, 1955.
14. Boyes JH: Why tendon repair? J Bone Joint Surg (Am) 41:577–
579, 1959.
15. Boyes JH: Bunnell’s Surgery of the Hand, ed 4, Philadelphia,
1964, JB Lippincott.
16. Boyes JH: The philosophy of tendon surgery. In AAOS Sym­
posium on Tendon Surgery in the Hand, St Louis, 1975, CV
Mosby, pp 1–5.
17. Boyes JH, Stark HH: Flexor tendon grafts in the fingers and
thumb: A study of factors influencing results in 1000 cases,
J Bone Joint Surg (Am) 53:1332–1342, 1971.
18. Bassett CAL, Carroll RE: Formation of tendon sheaths by
silicone rod implants, J Bone Joint Surg (Am) 45:884–885,
1963.
19. Hunter JM. Artificial tendons. Early development and appli­
cation, Am J Surg 109:325–338, 1965.
20. Hunter JM, Salisbury RE: Flexor-tendon reconstruction in
severely damaged hands. A two-stage procedure using a sili­
cone Dacron reinforced gliding prosthesis prior to tendon
grafting, J Bone Joint Surg (Am) 53:829–852, 1971.

279
CHAPTER
26 
EXPERIENCE WITH SECONDARY
FLEXOR TENDON REPAIRS
Peter C. Amadio, MD, and Chunfeng Zhao, MD
Between roughly 1920 and 1960, secondary repair was
the preferred choice when dealing with flexor tendon
injuries in zone 2.1-4 Subsequent advances in primary
repair, especially with newer suture designs and reha­
bilitation techniques, have made primary repair the pre­
ferred option for the vast majority of zone 2 injuries.
Nonetheless, secondary repair options remain viable in
many circumstances. Newer research may also expand
the role of secondary reconstruction, as described in the
final section of this chapter.
ONE-STAGE GRAFT
Indications
The classic one-stage tendon graft was described and
popularized by Bunnell,1 with definitive descriptions of
outcomes reported by Boyes, Stark, and others in the
middle years of the twentieth century.2,5,6 These descrip­
tions have not been bettered and serve as the reference
point here for a description of the classic indications,
technique, and results of one-stage grafting.
The classic indications for a one-stage graft are a zone
2 injury for which repair is not appropriate, with an
otherwise good tendon bed, in a cooperative patient.
These criteria remain valid; indeed, the only difference
between the 1950 and 2010 indications at all is the
assessment of which zone 2 injuries might not be appro­
priate for primary repair.
Another feature that has not changed over the years
is that it is often difficult to be certain beforehand
whether a tenolysis, one-stage graft, or two-stage graft
will be needed preoperatively; a procedure that begins
as a tenolysis may easily end with a shredded tendon,
damaged pulleys, and a two-stage reconstruction. The
surgical consent should include consideration of each
of these options, whenever tendon reconstruction is
discussed.
Currently, tendon injuries not suitable for primary
repair in zone 2 include neglected injuries, in which
retraction and adhesion of the tendon ends preclude
reapproximation, or only permit reapproximation
with extreme positioning of the finger and wrist in
flexion, and segmental injuries, especially with segmen­
tal tendon loss. In such cases primary repair is not
OUTLINE
Between roughly 1920 and 1960, secondary repair was
the preferred choice when dealing with flexor tendon
injuries in zone 2. Subsequent advances in primary
repair, especially with newer suture designs and reha­
bilitation techniques, have made primary repair the
preferred option for the most zone 2 flexor tendon
injuries. Currently, secondary repair options remain
viable in a number of circumstances. Secondary repair
in zone 2 include neglected injuries, in which retrac­
tion and adhesion of the tendon ends preclude reap­
proximation, or only permit reapproximation with
extreme positioning of the finger and wrist in flexion,
and segmental injuries, especially with segmental
tendon loss. In such cases primary repair is not techni­
cally possible. A good tendon bed is an indication for
one-stage grafting, such as when there is no or very
little scar, the pulleys (or at least the major pulleys A2
and A4) are intact, and there are no or minimal associ­
ated joint contractures. In addition, soft tissue nutri­
tion must be adequate, with preservation of at least
one neurovascular bundle. Finally, soft tissue coverage
must be normal or nearly normal: tendon grafts do
not thrive under scar or skin grafts. When tendon graft­
ing is indicated to restore tendon function and a one-
stage graft is relatively contraindicated because of
excessive scar, joint contracture, pulley loss, or a need
for better soft tissue cover, a two-stage reconstruction
can be considered. The patient should be fully informed
as to the time commitment to the staged surgery: the
initial stage, to address the scar, contracture, pulley and
soft tissue issues that contraindicate a one-stage graft;
an interval for soft tissue healing, which is usually 3
months or longer; a second stage, a “one-stage” graft;
and interval for graft healing; and the potential need
for a third stage, namely tenolysis or some sort of
salvage if the two-stage graft fails. Current research
efforts indicate improvement of the gliding character­
istics of extrasynovial autografts after tendon surface
modification with carbodiimide-derivatized hyalu­
ronic acid mixture and possibility of using tendon
allografts, which point to future promise of changes
in secondary tendon repair or reconstruction.

280
Section 3:  Secondary Flexor Tendon Surgery
quality of the soft tissues. The wound should be clini­
cally stable; the skin soft and pliable. If good soft tissues
are not present, the procedure is almost certain to
fail. Indeed, poor soft tissues are an indication for an
alternate procedure, such as a staged tendon graft, as
described next.
The finger can be approached through either a volar
Bruner or midaxial incision. The key decision is whether
to put the proximal juncture in the palm or the distal
forearm. This decision should be based on the quality
of the soft tissue bed; the palm is preferred provided
that the soft tissues there are soft and supple. Otherwise,
a longer graft, repaired proximal to the carpal tunnel, is
preferred.
At least under the pulleys, the old tendon should be
excised. If the pulleys are not intact, especially the key
A2 pulley, pulley reconstruction as a two-stage proce­
dure should be strongly considered.
The tendon graft donor should be considered care­
fully. Ideally, in an intrasynovial location, an intrasyno­
vial graft should be used, but these are rarely available.
In addition, to minimize bowstringing, a graft that fits
the pulley should be used, but again, a good diameter
match is rarely on offer. When such tendons are avail­
able, they should be considered. Some creativity can
sometimes fashion a better choice: an index extensor is
intrasynovial at the extensor retinaculum; such a tendon,
harvested and then reversed, so that the distal end is
proximal and vice versa, can serve as a good alternative
to a palmaris longus, when a short graft is needed.8
When a long graft is needed for a juncture in the distal
forearm, usually only a plantaris or toe extensor have
the required length. Alternatively, tendon allografts can
be used, but without some modification (see later) the
results may be suboptimal.
The proximal juncture is classically made with a Pul­
vertaft weave (Figure 26-2), but other weaving tech­
niques can provide similar or even superior holding
strength, with less bulk.9 There is little question, though,
technically possible. Even so, one-stage grafting might
not be indicated—in the presence of an intact and
functioning flexor digitorum superficialis (FDS) tendon,
grafting to restore flexor digitorum profundus (FDP)
function might cause more harm than good.6 It is always
important to recall that in the presence of a normal
FDS, the FDP contributes only to distal interphalangeal
joint (DIP) motion. Due to inevitable adhesions, the
amount of restored DIP motion is often limited, and
the amount of incremental grip strength may also be
limited. These limited gains must be counterbalanced by
the risks of infection and contracture, the time and effort
required for a successful tendon graft procedure, includ­
ing the possible need for a tenolysis, and the impact of
these on the patient’s life and livelihood. Except in
special circumstances, such as a musician who plays a
stringed instrument, serious consideration should be
given to simpler alternatives, such as DIP tenodesis, with
FDS tenolysis and lumbrical release as needed.
Certainly, the criterion of a good tendon bed has not
changed as an indication for one-stage grafting (Figure
26-1). A good bed is one in which there is no or very
little scar, the pulleys (or at least the major pulleys A2
and A4) are intact, and there are no or minimal associ­
ated joint contractures. In addition, soft tissue nutrition
must be adequate, with preservation of at least one
neurovascular bundle.5 Finally, soft tissue coverage must
be normal or nearly so: Tendon grafts do not thrive
under scar or skin grafts.
Finally, the patient must be capable of cooperating
with the grafting procedure. Sufficient voluntary motor
control must be available to power the graft motor, and
the patient must be able to cooperate with the rehabili­
tation program, which will include typically several
months of protected mobilization. Children, especially
under the age of 10, may not be fully capable of coop­
eration; reconstruction should be undertaken with
caution in such patients.7
METHODS
The technique of one-stage tendon grafting has not
changed over the years. Timing is dependent on the
Figure 26-1  A finger suitable for one-stage tendon graft:
tendon bed is good; scarring is minimal; the pulleys are
intact; and the soft tissues are supple.
Figure 26-2  The Pulvertaft weave is useful, especially to
join tendons of dissimilar sizes.

281
Chapter 26:  Experience With Secondary Flexor Tendon Repairs
first few weeks, to provide a limited range of graft
gliding, and then progressing along a rehabilitation
pathway similar to that for tendon repairs.
RESULTS
The results of one-stage grafting are best summarized in
the classic 1971 report by Boyes and Stark,5 which
reviewed their results in 1000 cases. It is unlikely that
anyone will ever duplicate this experience, so it is worth
reviewing. They classified the digits as being either good,
scarred, with joint contractures, as one of multiple
injured digits in the same hand, and as salvage cases, in
which there were associated skin, bone, or nerve inju­
ries. Results were graded based on active flexion; resid­
ual flexion contractures were not considered unless they
were more than 40° in the index and middle fingers and
60° in the ring and little fingers. According to this
scheme, the good digits did best, with two of three
flexing to within 0.5 inch to the palm, and 85% flexing
within 1 inch. The corresponding figures for the mul­
tiple and scar categories were a little worse: 30% and
60%, respectively. The digits with joint contractures
were worse again: 10% and 50%, respectively, while the
salvage cases did extremely poorly: no digits in this
group could flex with 0.5 inch and only 20% could flex
within 1 inch. Patients over age 30 did worse, and
patients over age 40 did much worse, for similar grades
of digit. So long as the digit remained supple, delay to
surgery did not seem to matter.
TWO-STAGE GRAFT
Indications
When tendon grafting is indicated to restore tendon
function and a one-stage graft is relatively contraindi­
cated because of excessive scar, joint contracture, pulley
loss, or a need for better soft tissue cover, then a two-
stage reconstruction can be considered. One aspect of
the indications that does not change between the one-
and two-stage procedure is the absolute need for a coop­
erative patient, who is fully informed as to the time
commitment that is likely to be necessary: the initial
stage, to address the scar, contracture, pulley and soft
tissue issues that contraindicate a one-stage graft; an
interval for soft tissue healing, which is usually 3 months
or longer; a second stage, which is a “one-stage” graft as
described earlier; and interval for graft healing, again as
described earlier; and then consideration of the need for
a possible third stage, namely tenolysis or some sort of
salvage if the two-stage graft fails. Only a patient willing
and able to travel this long road with the surgeon should
be considered for surgery; only a need great enough to
justify such an arduous journey should prompt the
surgeon to suggest it. Thus, for example, such proce­
dures are rarely indicated in young children7 or in those
who live at such a distance from the surgeon that close
that when joining tendons of grossly dissimilar diam­
eter, a weaving method is superior in strength to a direct
repair. The choice of donor motor depends to some
extent on what is available, but usually either the pro­
fundus or superficialis tendon of the affected digit is
chosen.
Distally, the graft is usually fixed with a pullout
suture. There is some controversy as to whether the
tendon graft should be drilled into bone or fixed to
the bony surface of the distal phalanx; recent research
suggests that surface fixation may result in better
healing.10 Tension may be adjusted proximally or dis­
tally; the operated finger should usually be slightly
more flexed than the normal cascade would suggest
(Figure 26-3).
Postoperatively, the patient is managed usually with
early passive mobilization, to reduce the risk of joint
contracture. The role of active motion is somewhat con­
troversial in tendon grafting; unlike a direct repair, the
graft is avascular, and must have its nutrition restored
somehow. Thin grafts may be able to survive on synovial
nutrition, but thicker grafts will almost certainly require
some adhesions to deliver a vascular source. The balance
that results is extremely delicate and likely explains the
usually mediocre results of tendon grafting—moving
too much and too soon disrupts adhesions and may
jeopardize long-term graft viability; protecting too long
will result in too many, short adhesions that unaccept­
ably limit function of a better vascularized graft. Our
compromise has been to use passive mobilization in the
Figure 26-3  Tendon cascade after repair; tension should
be set slightly tighter than normal.

282
Section 3:  Secondary Flexor Tendon Surgery
and when they break they are difficult to reconstruct.
Proximal failure typically leaves a destroyed muscle–
tendon junction, which is hard to salvage.
After stage 1, the finger is passively mobilized as the
wound conditions permit. Stage 2 is performed when
the incisions are soft and supple and passive finger
motion has reached a maximum. Stage 2 surgery is a
tendon graft, repaired and rehabilitated as for one-stage
grafts. The main difference is that often longer grafts are
needed, such as plantaris or a toe extensor.
RESULTS
The results of two-stage grafts are certainly better than
those of a one-stage graft performed for similar indica­
tions but remain unexciting in many cases.13-19 Flexion
contractures and limited flexion remain common; the
recovery period is long, and patient satisfaction is often
low, with expectations incompletely met. Among the
more common complications are graft rupture and late
contracture, especially of the DIP joint. This latter com­
plication can be avoided if the graft is used to recon­
struct the flexor superficialis, and the distal joint is
fused.
SECONDARY REPAIR IN THE FUTURE
Improving the gliding characteristics of extrasynovial
autografts may lead to improved clinical results
after tendon graft surgery. Recent studies have shown
that extrasynovial graft surface modification with
carbodiimide-derivatized
hyaluronic
acid
mixture
decreased gliding resistance of the tendon and
increased the durability after repetitive motion in the
in vitro studies.20,21 Zhao et al conducted an in vivo
flexor tendon graft in canine model using autologous
peroneus longus tendon. They demonstrated that the
surface of an extrasynovial tendon autograft with
a carbodiimide-derivatized hyaluronic acid gelatin
decreases adhesions and increases tendon gliding.22
These encouraging experimental results potentially
provide surgeons with a new and useful method to
improve the quality of tendon graft surgery.
Although intrasynovial tendon autologous grafts
are rarely obtained without compromising the donor
site morbidity, allograft-matched intrasynovial flexor
tendons are possibly available. Allograft tendon grafts
are frequently used for anterior cruciate ligament recon­
struction,23-27 but flexor tendon reconstruction with
allograft tendon has been rarely reported.28 The use
of allografts for tendon reconstruction merits serious
consideration for the following reasons: intrasynovial
allograft tendon sources are abundant, which permits
easy clinical application; allograft use limits surgical
morbidity, as no graft harvest is necessary; allograft use
reduces the surgical time required by eliminating the
graft harvesting procedure, which directly translates into
a cost reduction29; and better size matching. A major
follow-up becomes impossible.11 Reconstruction of a
profundus tendon in the presence of a normally func­
tioning superficialis must be carefully considered—Will
the potential benefit outweigh the considerable risks6?
Even reconstruction of the thumb flexor should be care­
fully thought through—How does the risk/benefit ratio
compare, for example, with a much simpler interpha­
langeal arthrodesis?
METHOD
The first-stage procedure involves addressing the soft
tissue issues outlined earlier and then using a silicone
rubber spacer to maintain a place for the second-stage
tendon graft (Figure 26-4). Thus, each operation is
unique. Nevertheless, some basic principles are useful
to note. Joint contractures should be released prior to
soft tissue reconstruction, so that the full length of
needed tissue can be ascertained. When releasing con­
tractures in the presence of pulley loss, it is especially
important to excise all scar anterior to the bone; often
a hard triangle of scar will feel like bone and may deceive
the unwary surgeon, who then reconstructs a pulley that
does not hold the graft close enough to the bone to
prevent postoperative bowstringing.
As with one-stage grafting, a key decision is where
to put the proximal junction—in the palm or forearm.
In the case of one-stage grafts, which ought to have
good surrounding tissues, usually the palm is appropri­
ate. In the case of two-stage grafts, usually it is the
opposite—the palm and the finger are scarred, and the
most hospitable location is often the distal forearm.
Regardless, the tendon implant should be inserted and
fixed distally to bone, with either screw or suture. It is
threaded through any intact pulleys and pulleys are then
reconstructed around the graft where needed. When
completed, the graft should fit snugly within the new
sheath but should piston smoothly, without any ten­
dency for buckling with passive flexion. Any buckling at
this stage will simply become synovitis and infection
postoperatively.
We do not use active implants.12 The literature sug­
gests that their results are no better than passive ones,
Figure 26-4  Stage 1 surgery with a silicone implant in
place. Note pulley reconstruction at the middle part of the
proximal phalanx.

283
Chapter 26:  Experience With Secondary Flexor Tendon Repairs
function, reduced adhesions, and decreased gliding
resistance compared with allografts without such modi­
fication (Figure 26-7).36 We believe that treated allografts
may similarly improve the results after flexor tendon
reconstruction in humans.
Figure 26-5  Lubricated tendon graft after chemical
modification of tendon surface.
Figure 26-6  Scarred digit after initial surgery in an animal
model of tendon grafting.
Figure 26-7  Treated (top) and untreated (bottom) grafts in
a canine model. Adhesions were markedly reduced in the
treated graft.
disadvantage is the potential for an immune reaction,
but this can reduced by processing the allograft prior to
use. A lyophilized, decellularized allograft is less immu­
nogenic, better tolerated, and as effective as autograft.30-32
Preserved allografts also maintain their mechanical
properties with the convenience of availability at any
desired time.25,33,34
In addition to the needs of mechanical strength and
appropriate tendon/bone healing, the flexor tendon
allograft also requires a smooth gliding surface. However,
Ikeda et al found that the procedures used to prepare
allografts damage the tendon surfaces leading to greater
frictional force in canines.35 This adverse effect can
be reversed by surface treatment with carbodiimide-
derivatized hyaluronic acid and gelatin (Figure 26-5).35
A flexor tendon reconstruction with surface modified
allografts has been studied in a canine in vivo model of
failed primary repair.36 In this experiment, the flexor
tendons were lacerated and repaired, and the animals
were allowed free active motion postoperatively. The
repaired tendons were all ruptured within 2 weeks. After
6 weeks, a scarred digit with a uniform failure pattern
was routinely produced (Figure 26-6). Subsequent
grafting with surface-treated allografts improved digital
References
1. Bunnell S: Hand surgery, J Bone Joint Surg (Am) 29:824,
1947.
2. Boyes JH: Evaluation of results of digital flexor tendon grafts,
J Surg (Am) 89:1116–1119, 1955.
3. Hauge MF: The results of tendon suture of the hand: A review
of 500 patients, Acta Orthop Scand 24:258–270, 1955.
4. Kelly AP Jr: Primary tendon repairs: A study of 789 consecu­
tive tendon severances, J Bone Joint Surg (Am) 41:581–598,
1959.
5. Boyes JH, Stark HH: Flexor-tendon grafts in the fingers and
thumb. A study of factors influencing results in 1000 cases,
J Bone Joint Surg (Am) 53:1332–1342, 1971.
6. Stark HH, Zemel NP, Boyes JH, et al: Flexor tendon graft
through intact superficialis tendon, J Hand Surg (Am) 2:456–
461, 1977.
7. Amadio PC: Staged flexor tendon reconstruction in children,
Ann Chir Main Memb Super 11:194–199, 1992.
8. Nishida J, Amadio PC, Bettinger PC, et al: Excursion
properties of tendon graft sources: interaction between
tendon and A2 pulley, J Hand Surg (Am) 23:274–278,
1998.
9. Tanaka T, Zhao C, Ettema AM, et al: Tensile strength of a new
suture for fixation of tendon grafts when using a weave
technique, J Hand Surg (Am) 31:982–986, 2006.

284
Section 3:  Secondary Flexor Tendon Surgery
25. Tejwani SG, Shen W, Fu FH: Soft tissue allograft and
double-bundle reconstruction, Clin Sports Med 26:639–660,
2007.
26. Dustmann M, Schmidt T, Gangey I, et al: The extracellular
remodeling of free-soft-tissue autografts and allografts for
reconstruction of the anterior cruciate ligament: A compari­
son study in a sheep model, Knee Surg Sports Traumatol
Arthrosc 16:360–369, 2008.
27. Scheffler SU, Schmidt T, Gangéy I, et al: Fresh-frozen
free-tendon allografts versus autografts in anterior cruciate
ligament reconstruction: delayed remodeling and inferior
mechanical function during long-term healing in sheep,
Arthroscopy 24:448–458, 2008.
28. Liu TK: Clinical use of refrigerated flexor tendon allografts to
replace a silicone rubber rod, J Hand Surg (Am) 8:881–887,
1983.
29. Cole DW, Ginn TA, Chen GJ, et al: Cost comparison of
anterior cruciate ligament reconstruction: Autograft versus
allograft, Arthroscopy 21:786–790, 2005.
30. Webster DA, Werner FW: Mechanical and functional proper­
ties of implanted freeze-dried flexor tendons, Clin Orthop
Relat Res 180:301–309, 1983.
31. Gulati AK, Cole GP: Nerve graft immunogenicity as a factor
determining axonal regeneration in the rat, J Neurosurg
72:114–122, 1990.
32. Fromm B, Schäfer B, Parsch D, et al: Reconstruction of the
anterior cruciate ligament with a cyropreserved ACL allograft.
A microangiographic and immunohistochemical study in
rabbits, Int Orthop 20:378–382, 1996.
33. Goertzen MJ, Clahsen H, Schulitz KP: Anterior cruciate liga­
ment reconstruction using cryopreserved irradiated bone-
ACL-bone-allograft transplants, Knee Surg Sports Traumatol
Arthrosc 2:150–157, 1994.
34. Mahirogullari M, Ferguson CM, Whitlock PW, et al: Freeze-
dried allografts for anterior cruciate ligament reconstruction,
Clin Sports Med 26:625–637, 2007.
35. Ikeda J, Zhao C, Sun YL, et al: Carbodiimide-derivatized hyal­
uronic acid surface modification of lyophilized flexor tendon:
A biomechanical study in a canine in vitro model, J Bone Joint
Surg (Am) 92:388–395, 2010.
36. Zhao C, Sun YL, Ikeda J, et al: Improvement of flexor tendon
reconstruction with carbodiimide-derivatized hyaluronic
acid and gelatin-modified intrasynovial allografts: Study of
a primary repair failure model, J Bone Joint Surg (Am) 92:
2817–2828, 2010.
10. Silva MJ, Thomopoulos S, Kusano N, et al: Early healing of
flexor tendon insertion site injuries: Tunnel repair is mechan­
ically and histologically inferior to surface repair in a canine
model, J Orthop Res 24:990–1000, 2006.
11. Amadio PC, Wood MB, Cooney WP 3rd, et al: Staged flexor
tendon reconstruction in the fingers and hand, J Hand Surg
(Am) 13:559–562, 1988.
12. Hunter JM, Singer DI, Jaeger SH, et al: Active tendon implants
in flexor tendon reconstruction, J Hand Surg (Am) 13:849–
859, 1988.
13. Sun S, Ding Y, Ma B, et al: Two-stage flexor tendon reconstruc­
tion in zone II using Hunter’s technique, Orthopedics 33:880,
2010.
14. Alnot JY, Masmejean EH: The two-stage flexor tendon graft,
Tech Hand Up Extrem Surg 5:49–56, 2001.
15. Smith P, Jones M, Grobbelaar A: Two-stage grafting of flexor
tendons: results after mobilisation by controlled early active
movement, Scand J Plast Reconstr Surg Hand Surg 38:220–227,
2004.
16. Amadio PC, Wood MB, Cooney WP 3rd, et al: Staged flexor
tendon reconstruction in the fingers and hand, J Hand Surg
(Am) 13:559–562, 1988.
17. Wehbé MA, Mawr B, Hunter JM, et al: Two-stage flexor-
tendon reconstruction. Ten-year experience, J Bone Joint Surg
(Am) 68:752–763, 1986.
18. Schneider LH: Staged flexor tendon reconstruction using the
method of Hunter, Clin Orthop Relat Res 171:164–171, 1982.
19. Paneva-Holevich E: Two-stage reconstruction of the flexor
tendons, Int Orthop 6:133–138, 1982.
20. Momose T, Amadio PC, Sun YL, et al: Surface modification
of extrasynovial tendon by chemically modified hyaluronic
acid coating, J Biomed Mater Res 59:219–224, 2002.
21. Sun YL, Yang C, Amadio PC, et al: Reducing friction by chemi­
cally modifying the surface of extrasynovial tendon grafts,
J Orthop Res 22:984–989, 2004.
22. Zhao C, Sun YL, Amadio PC, et al: Surface treatment of flexor
tendon autografts with carbodiimide-derivatized hyaluronic
acid. An in vivo canine model, J Bone Joint Surg (Am) 88:2181–
2191, 2006.
23. Jackson DW, Grood ES, Goldstein JD, et al: A comparison
of patellar tendon autograft and allograft used for anterior
cruciate ligament reconstruction in the goat model, J Sports
Med (Am) 21:176–185, 1993.
24. Kustos T, Bálint L, Than P, et al: Comparative study of auto­
graft or allograft in primary anterior cruciate ligament recon­
struction, Int Orthop 28:290–293, 2004.

285
OUTLINE
Primary repair of the flexor pollicis longus tendon is
generally recommended, except in zone 3 or 4 injuries,
when primary grafting can be advisable. Techniques
are discussed that allow primary repair under circum­
stances that might otherwise require tendon grafting.
Circumstances in which no secondary repair is accept­
able are also discussed. The requisites for function that
must be met if the flexor pollicis longus muscle is to
be used as the motor for a tendon graft are examined.
Some technical aspects of tendon grafting are dis­
cussed and the usefulness of two-stage grafting under
certain circumstances is considered.
The technical execution of the various surgical maneu­
vers used in secondary reconstruction of the flexor pol­
licis longus (FPL) (Box 27-1) are relatively straight
forward to any hand surgeon versed in the techniques
of flexor tendon surgery. However, decision making as
to what is done in any particular circumstance is com­
plicated by various anatomical peculiarities of the FPL
tendon compared to the flexor system of the finger, the
timing of presentation, the plethora of treatment
options, and, in some instances, by the differing flexion
needs of the thumb in different individuals.
PRIMARY REPAIR OF FLEXOR POLLICIS
LONGUS BY PREFERENCE
Recent practice (see Chapter 16) has established that
primary repair of the FPL tendon is desirable where pos­
sible and 70% to 80% of the results will be within the
excellent or good range, although it should be noted
that “excellent” in our current classifications does not
imply return to normal function. Nevertheless, these
results are better than the 30° to 40° degrees of move­
ment previously perceived as necessary to good function
of the interphalangeal (IP) joint of the thumb.1,2 With
various modern suturing techniques of primary repair,
early mobilization can also be progressed safely to com­
pletion without rupture of the repair.3,4 Direct repair
CHAPTER
27 
SECONDARY RECONSTRUCTION
OF THE FLEXOR POLLICIS
LONGUS TENDON
David Elliot, MA, FRCS, BM, BCh
may be possible as late as 3 to 4 weeks after injury but,
more frequently, in my experience, there is sufficient
shortening of the muscle within a few days as to make
direct primary repair without proximal tendon length­
ening impossible without the IP being flexed to a degree
from which it never recovers extension.
PRIMARY GRAFTING OF ZONE 3 AND
4 INJURIES
The FPL is rarely divided in zone 3, within the thenar
muscles, where it is deeply placed. However, surgical
repair is difficult in zone 3 as access is past the digital
nerves of the thumb and through the thenar muscles,
with risk to the motor branches of the median nerve.
That loss of thumb sensation or opposition are greater
disabilities than loss of terminal flexion should be con­
sidered before extending an already extensive injury in
this zone to repair the tendon directly.5 Any direct
tendon repair in zone 3 will also be surrounded by
edematous and swollen muscle during rehabilitation.
Except where the injury is a small puncture wound of
the thenar eminence and dissection to the site of divi­
sion can be precise and small, there is justification for
the approach recommended by Matev (1983) of replac­
ing the tendon from the wrist to its distal attachment
with a tendon graft, even if seen immediately.6 In these
injuries, the proximal end of the tendon is frequently
found retracted to the wrist. When presentation is even
only slightly delayed, the distal end of the tendon may
be adherent within the muscle tunnel and require
careful release from adhesions through the digital expo­
sure, if necessary leaving the end itself within the muscles
by division of the tendon just proximal to the metacar­
pophalangeal joint.
Isolated division of FPL in the carpal tunnel (zone 4)
is even rarer. Direct tendon repair in the carpal tunnel
may snag on the relatively narrow opening from the
carpal tunnel into the tunnel through the thenar muscles
or on the proximal edge of the carpal ligament. Under
these circumstances, the Matev approach of grafting the
FPL from the wrist to the distal phalanx of the thumb

286
Section 3:  Secondary Flexor Tendon Surgery
extension of primary repair or a technique of secondary
reconstruction. This has been discussed in some detail
in Chapter 16 but more from the point of view of avoid­
ing a tightly flexed interphalangeal (IP) joint of the
thumb and/or tendon rupture by primary suture under
tension.
The considerable literature of the 1940–1960 period,
when most presentations and repairs were delayed,
presents a conflict of opinion as to whether the FPL
should be repaired by interpositional grafting or by
proximal tendon lengthening, either in the muscle or
in the tendon at the wrist, with direct repair being
entirely avoided by many authors, or exceptional, except
after very early presentation for others. Urbaniak and
Goldner favored direct repair when possible and used
the other techniques when the tendon gap was too
wide.10,11 They found the results of tendon lengthening
in their unit to be better than those of interposition
grafting in those cases in which direct repair was not
possible. In their writings, others are more circumspect
as to whether tendon lengthening gives better results
than tendon grafting.5,6 This debate as to the relative
merits of extending primary repair by proximal length­
ening or moving on to grafting continued through the
period after 1970 and remains in the literature with
Schneider (1999), in Green’s Operative Hand Surgery,
confessing to little experience with tendon lengthen­
ing,12 while Matev (1983) considered Z lengthening in
the musculotendinous part of the tendon as a logical
extension of primary suturing and preferable to grafting
when the muscle has retracted but not undergone
degeneration and fibrosis.6 Our experience has largely
been after only slight delay of presentation, with muscle
shortening sufficient for primary suture to be under
tension, or significantly flex the IP joint, but before
muscle degeneration. Proximal tendon lengthening
within the muscle by the Le Viet technique13 has usually
achieved the lengthening of up to 1 cm required in
these circumstances. My experience of “Z” lengthening
of the FPL tendon at the musculotendinous junction14,15
is limited and usually as an adjunct to lengthening
within the muscle where this alone is not adequate
(Figure 27-2). Pulvertaft (1966) wrote that a gap of 112
to 2 inches (3.8 to 5 cm) could be closed by this tech­
nique alone.5 The long length of FPL tendon—3 to
4 cm—at the wrist with no muscle fibers attached aids
this degree of lengthening without impingement of the
tendon sutures on the carpal ligament on full extension
of the thumb with the wrist dorsiflexed, if the lengthen­
ing is carried out at the musculotendinous junction.10
The adjacent muscle can also be tacked around the
lengthening to make it more smooth. I have no experi­
ence of Rouhier lengthening within muscle, still said to
be used in France.16,17
The practice of advancing the proximal end of the
FPL to allow direct reattachment to the distal phalanx
also has some merit. With multiple tendon injuries
within the carpal tunnel, swelling of the tendons makes
tethering of the multiple repairs within the tunnel likely.
With this rare injury, we have previously provided suf­
ficient space to avoid the problem of tethering of the
swollen tendons and impeded movement by repairing
only the finger profundus tendons and have repaired
the FPL tendon primarily. To achieve the necessary
repairs in the carpal tunnel requires division of the
carpal ligament. To allow for swelling and avoid possi­
ble compression of the median nerve, we have not
reconstructed the carpal ligament but have mobilized
the tendons repairs in a neutral, or slightly hyperex­
tended, wrist position.
TECHNIQUES TO EXTEND PRIMARY REPAIR
In general, the use of primary flexor surgery can be
extended and secondary surgery avoided by (1) using
techniques that allow one to do more primary repairs,
such as undertaking delayed primary repairs, using
proximal tendon lengthening, and using techniques
such as splitting swollen tendons distally to allow their
passage through the pulleys7; (2) by using techniques
that reduce the failures of primary repair, such as
improved rehabilitation; and (3) by re-repairing rup­
tures.8 Many of these techniques are applicable to the
divided FPL tendon. Occasionally, the FPL may be too
swollen to pass through the A1 and A2 pulleys. The
technique of halving the distal end of the tendon that
we described to pass a swollen flexor digitorum profun­
dus (FDP) tendon through the A4 pulley7 can also be
used in FPL surgery (Figure 27-1). However, the most
significant problem preventing primary FPL repair is
the propensity of the FPL muscle to contract after divi­
sion of the tendon,9 and proximal tendon lengthening
to deal with this eventuality can be considered an
Treatment Options for the Divided
FPL Tendon
Primary and Delayed Primary Repair
1. End-to-end suture of the cut tendon
(Advised in all but zone 3)
2. Reattachment of the distal tendon to the distal phalanx
(a) After tendon avulsion
(b) After division close to the distal insertion
Extended Primary Repair
Primary repair with proximal tendon lengthening
Secondary Reconstruction
1. No surgery
2. Distal tenodesis of the FPL or IP joint fusion
3. Tendon grafting in one stage
4. Tendon grafting in two stages
5. Tendon transfer
Box 27-1